Estrogen, menopause and joints

A spectrum of musculoskeletal symptoms occurs at menopause, ranging from arthralgia to osteoarthritis¹. In fact, joint aches are some of the most commonly reported symptoms at menopause. These typically affect the small joints of the hand and can give rise to pain and impaired function. There are clear sex-specific differences in the biology of osteoarthritis (OA), with increasing prevalence in women after menopause leading to twice as many women affected as men. Typically the OA will affect the distal interphalangeal joints of the hands, with or without involvement of the proximal interphalangeal and carpometacarpal joints. Involvement of these joints can occur without evidence of OA at other sites. The International Menopause Society has previously highlighted the significant impact of OA in women². However, OA is a particularly 'unsexy' area of medicine so, along with the general lack of understanding of its pathogenesis and the apparent absence of any disease-modifying treatment, we often see these symptoms being disregarded.

Any real understanding of the pathogenesis of OA is still in its infancy. Family history, body mass index, trauma and over-use all have a role to play. However, gender differences in its incidence would also suggest a role for sex steroids. Data collected from a multidisciplinary hand clinic in London showed that 80% of the patients were women and the majority had developed their symptoms within 4 years of reaching menopause or stopping their menopause hormone therapy (MHT)³. This association has also been seen in young women with an early menopause and in those taking aromatase inhibitors after breast cancer⁴.

If estrogen does indeed slow OA progression and reduce pain and stiffness, how is this mediated? OA is a disease that affects the entire joint and we know that estrogen receptors (ERs) are present in cartilage, sub-chondral bone, synovium, muscles and tendons. Deletion of ERs in animal models causes cartilage damage and changes in the sub-chondral bone, but studies of ERα polymorphisms and OA have produced conflicting results^5,6. There are also conflicting data on the link between endogenous sex steroid levels and the risk of OA. Estrogen appears to have pro- or anti-inflammatory effects depending on the situation^7,8. Loss of estrogen increases reactive oxygen species and pro-inflammatory cytokines. In animal models, estrogen and androgens, but not progesterone, appear to attenuate the effects of ovariectomy. Animal studies examining the effect of tibolone after induced joint damage have also demonstrated its ability to reduce cartilage degeneration and synovitis. Replacing estrogen appears to have sex-specific benefits limited only to females. Incontrast to these positive findings, a number of groups have demonstrated pro-inflammatory effects on cartilage and sub-chondral bone with high-dose estrogen. So, perhaps dose is important.

Aside from basic science models, the critical question for clinicians is whether MHT provides any benefit in postmenopausal women. Here, once again, there are mixed data. Some studies have shown no effect, while others have shown a protection from estrogen therapy. The Ulm Osteoarthritis Study reported no significant association between MHT use and radiographic hip, knee, or hand OA in 475 women⁹. However, women from the Study of Osteoporotic Fractures who were currently using oral estrogen had a significantly reduced risk of OA of the hip or moderate-to-severe radiographic manifestations of OA. Current users who had taken estrogen for at least 10 years had a greater reduction in the risk of any OA of the hip compared with that of users for less than 10 years¹⁰. The Women’s Health Initiative demonstrated lower rates of hip joint replacement in Caucasian women taking unopposed estrogen and lower rates of joint pain and stiffness in the combined hormone therapy arm¹¹. The latter was confirmed in the WISDOM trial¹². The reduction in pain has been hypothesized to occur via estrogen’s effect on the opioid system as well as direct effects on bone and cartilage. The General Practice Research Database retrospective cohort study demonstrated a 40% reduction in revision rates after knee and hip joint replacement surgery if estrogen was started following the arthroplasty¹³. Pre-operative use had no benefit, possibly because of suppression of bone resorption and issues with integration of the prosthesis.

Although in vivo, in vitro, genetic and clinical studies are assisting our understanding, without an intervention trial specially addressing hormones and joint health, the role of sex steroids remains open to debate. Advocating MHT as first-line treatment in postmenopausal women with OA is not where the data take us at the moment, particularly if long-term therapy is required. Questions that need to be answered include:

• Is estrogen a major contributor to joint health?

• Is any reduction in OA due to a protective effect of estrogen alone or do androgens or progestogens/progesterone play a role?

• What impact does the timing of MHT use have on OA progression or symptom severity?

• What impact does MHT have on outcomes from arthroplasty procedures?