Using a large research database representing all population groups in the United Kingdom, he and his colleagues examined the records of women undergoing a first hip or knee replacement between 1986 and 2006. Of the 24,733 women who had joint surgery, they selected 2,700 who had used HRT for at least six months before or after the surgery and matched them with 8,100 nonusers. The two groups were followed for at least three years.

At the end of the study, 168 patients had undergone joint revision surgery. After taking into account factors such as age, the joint replaced, use of other drugs (including calcium, vitamin D, corticosteroids and bisphosphonates) and existing medical conditions results showed that women who used HRT for at least six months after the original joint replacement were 38 percent less likely to need a repeat procedure than nonusers were. And those who continued taking HRT for a year were 50 percent less likely to undergo a second surgery. Taking medication for more than a year on a consistent basis (defined as at least 80 percent of the time) reduced the risk of implant failure even more. 

On the other hand, using HRT for any length of time before surgery had no effect on bone loss or implant failure. Dr. Arden suggests that presurgical HRT might actually do harm by preventing implants from anchoring properly to the bone. And he notes that hormone therapy should be stopped six weeks before surgery to reduce blood clot risk.

Blood clots are just one of the problems sometimes associated with HRT. The 2002 Women's Health Initiative (WHI) trial reported a slightly increased risk of breast cancer and heart disease in women who use hormone therapy for more than five years, although the applicability of those findings have since been questioned.

Dr. Arden points out that his study adds “a relevant piece of information for women who have received a total knee or hip replacement and are considering hormone replacement therapy for menopause. [But] it is only proof of principle and shouldn’t actually influence prescribing until [a randomized control trial] is performed.”

Elena Losina, PhD, codirector of the Orthopedics and Arthritis Center for Outcomes Research at Brigham and Women’s Hospital in Boston, calls the Oxford study “very well designed and executed using appropriate, modern analytic methods.”

But she adds, “As [the study authors] highlighted, in order to consider these results more definitively in clinical practice, they need to be confirmed and reproduced in a multicenter randomized controlled trial.”