Rheumatoid arthritis (RA) is clearly predominant in women, particularly before the age of 50. There are also peaks in incidence during the perimenopausal period, when ovarian function is declining. Disease activity also tends to increase during this period, but also during pregnancy, postpartum or breastfeeding. These notions have led to the search for an association between endogenous female hormones and the pathogenesis or prognosis of RA.
Conflicting studies
Several studies, mainly cross-sectional, often of the case-control type, have produced conflicting results regarding exposure to hormonal treatments: some have argued in favour of a protective effect of oral contraceptives (OCs), others have concluded the opposite, the latter being in favour of neutrality.
Since OCs are rarely prescribed in cases of RA, a reverse causal link has even been suggested to explain the potential beneficial effect of the latter, the lack of statistical power of certain studies being able to intervene in their inconclusive or inconclusive results, as well as approximate definitions of the disease.
The role of hormone replacement therapy (HRT) has been less widely published, but overall, their results are equally discordant, for similar reasons. HRT may have potential for preventing late-onset RA in older adults, known as LORA (late-onset RA) which begins abruptly and has a more severe impact on quality of life than the form in younger women.
In this regard, a South Korean retrospective case-control study of 6,056 cases of RA suggests that HRT (with a duration of exposure of 5 years) somewhat increases the risk of RA. These results are not contradicted by those of another study involving a cohort of older women.
A new study based on data from the UK Biobank
CO and HRT could therefore have opposite effects on the risk of RA: a hypothesis supported by the results of a new retrospective cohort study also carried out using data from the UK Biobank.
Two groups were formed, according to age and nature of treatment: OC (n = 236,602) and THS (n= 102,466). The associations between these hormonal treatments and the risk of RA were studied using the Cox proportional hazards model, with adjustments taking into account as many potential confounding factors as possible.
The diagnosis of RA was made according to the criteria of the International Classification of Diseases (10th edition), most often from hospitalization records. Sensitivity analyses were also conducted in both cases.
Long-term exposure to CO (versusno exposure) was associated with a reduced risk of RA, with an estimated hazard ratio (HR) of 0.89 (95% CI 0.82–0.96). The same was true for current (HR = 0.81; 95% CI 0.73–0.91) or previous (HR = 0.92; 95% CI 0.84–1.00) exposure. The same approach led to inverse results for HRT. Thus, an increased risk of RA, in its LORA form, was associated with long-term exposure to HRT (HR = 1.16; 95% CI 1.06–1.26) and the same was true for previous exposure (HR = 1.13; 95% CI 1.03–1.24).
This retrospective cohort study supports certain current hypotheses, without confirming them, given its methodology: OCs would be associated with a lower risk of RA, while HRT would have the opposite effect. No definitive conclusion can be envisaged in light of these results, which are added to those of previous methodologically similar studies. Only a prospective approach would allow a conclusion, which is difficult to implement in such a context.
