番茄社区

Nearly a decade of additional research on hormone replacement therapy (HRT) for preventing chronic diseases still indicates more risk than benefit, a new evidence review suggested.

Conducted to inform a new set of U.S. Preventive Services Task Force (USPSTF) recommendations on hormone therapy, the review by Heidi Nelson, MD, MPH, of Oregon Health and Science University in Portland, and colleagues, found that every current HRT regimen carries significant risks that, for most women, probably outweigh the benefit.

"Both hormone therapy regimens [estrogen unopposed or combined with progestin] decrease risk for fractures but increase risk for stroke, thromboembolic events, gallbladder disease, and urinary incontinence," they wrote online in Annals of Internal Medicine.

They added that "estrogen plus progestin also increases risk for breast cancer and probable dementia, whereas estrogen alone decreases risk for breast cancer."

In 2002, the USPSTF decided the risks of estrogen plus progestin outweighed likely benefits for preventing chronic conditions. The task force recommended against routine, long-term HRT use. It followed with a similar recommendation in 2005 for unopposed estrogen.

Although the recommendations did not address short-term use to relieve menopause symptoms, many physicians and their patients took them to mean that HRT was too dangerous for any purpose.

Since the previous USPSTF recommendations were published, many more studies of hormone therapy have appeared. The group decided it was time for an update and commissioned Nelson and colleagues to review the recent research from 2002 to 2011.

In particular, the researchers focused on randomized, placebo-controlled studies of menopausal HRT that evaluated primary prevention of conditions such as cardiovascular disease, cognitive decline, osteoporosis, and cancer.

They ultimately identified nine trials, all of at least fair quality, that merited inclusion.

Among them were:

• Two main trials in the Women's Health Initiative (WHI)
• Two spinoff WHI studies on memory and cognitive aging
• Estrogen Memory Study (EMS)
• Estrogen in the Prevention of Reinfarction Trial (ESPRIT),
• Ultra-Low-Dose Transdermal Estrogen Assessment (ULTRA)
• Women's International Study of Long-Duration Oestrogen after Menopause (WISDOM)
• Heart and Estrogen/Progestin Replacement Study (HERS)

The trials were too different in design and outcomes assessed to permit pooling of data. Instead, Nelson and colleagues wrote narrative summaries of what the studies found in answer to three key questions: what are the potential benefits of hormone replacement for preventing chronic conditions?, what are the harms?, and do benefits and harms differ by subgroups?

The researchers needed only three paragraphs to report the potential benefits, but they were clinically significant.

In the WHI, invasive breast cancer incidence and mortality were reduced by 23% and 63%, respectively, with unopposed estrogen (both P<0.05). Diabetes was reduced by 21% and 35% (both P<0.05) in the WHI and HERS trials with estrogen plus progestin, but unopposed estrogen in the WHI study did not have such an effect.

Hip, vertebral, and total fractures were reduced significantly with both hormone regimens in WHI but not in HERS. The WHI also held a possible signal for colorectal cancer prevention with estrogen plus progestin but it barely reached statistical significance (HR 0.75, 95% CI 0.57 to 1.00). That result was not confirmed in HERS, nor was it seen for unopposed estrogen in the WHI.

Among the harms that were significantly increased with one or both hormone replacement regimens:

• Invasive breast cancer (estrogen plus progestin)
• Stroke (both)
• Deep vein thrombosis (DVT, both)
• Pulmonary embolism (estrogen plus progestin)
• Breast cancer mortality (estrogen plus progestin)
• Lung cancer mortality (estrogen plus progestin)
• Gallbladder disease (both)
• Probable dementia (estrogen plus progestin)
• Urinary incontinence (both)

For both regimens, the number of women who would be expected to suffer harm per 10,000 person-years was far greater than those who would benefit from either regimen.

With unopposed estrogen, there would eight fewer invasive breast cancers, 56 fewer fractures, and two fewer breast cancer deaths, but 11 more strokes, seven more DVT episodes, 33 more gallbladder disease cases, and 1,271 cases of urinary incontinence per 10,000 person-years.

The corresponding numbers for estrogen plus progestin were just as unfavorable.

Nelson and colleagues were unable to answer their third key question, about special benefits or harms in subgroups. They had thought they might see different effects by age, type of hormone therapy, presence of comorbidities, or natural versus premature menopause.

But despite the large numbers of women in the trials -- more than 27,000 in the two main WHI trials, for example -- the subgroup analyses were "limited and inconclusive," Nelson and colleagues wrote.

The analysis as a whole had several limitations, the researchers noted. Most outcomes of interest were reported in no more than two trials. Low retention and adherence to protocols were problems in most of the trials.

In addition, the vast majority of trial participants were in their 60s, limiting applicability to women in menopause or immediately postmenopausal. Also, most of the trials used oral conjugated equine estrogens, raising questions about the effects of estrogens from other sources and delivered in other ways, such as transdermally.

No timetable was given for release of the updated USPSTF recommendations.

Comment