Rheumatoid arthritis in male patients : sex hormones, bone mineral density and clinical characteristics

The aims of this study were to investigate sex hormones in men with RA and the relation of these hormones with disease characteristics including bone mineral density, as well as to study gender differences in rheumatoid arthritis (RA). Estrogens and androgens and the balance between these hormones interact with the cytokine network and modulate immunity. Sex hormones are thus important for the phenotype of autoimmune diseases such as RA and hence for differences in disease incidence and severity between genders.

The work of this thesis is based on cross-sectional observational studies and longitudinal observational data. For the investigations on sex hormones in men with RA 104 men attending the Rheumatology department were examined. A control population was collected in the same region. To analyse gender differences, 844 patients with early RA followed prospectively in the multicentre "BARF0T-registry" were used. In both the local patient group and the larger national patient group disease activity was estimated by a disease activity score including tender and swollen joint count in 28 joints (DAS28), patient's assessment of global health and ESR. Physical disability was estimated by using the Health Assessment Questionnaire (HAQ).

Men with RA had several hormone aberrations. Bioavailable testosterone (NST) was in average lower in patients than in controls as were the adrenal androgen DHEAS and the estrogen derived from adrenal androgens - estrone. NST correlated with HAQ score i.e. patients with more physical impairment had lower NST, but with no other disease specific variable. DHEAS was lower in rheumatoid factor (RF) negative than RF positive patients, whereas no other variable differed between these two groups. NST and DHEAS did not correlate with markers of inflammation when calculated on the whole population, but when analysing only RF negative men these androgens had significant and inverse correlation with inflammation.

Estradiol levels were in average higher in the RA men than in controls and contrasting to the three former hormones the correlations in the whole population between markers of inflammation and estradiol were conspicuous. Increased conversion of estrone to estradiol seemed to explain a major part of the high estradiol levels.

Bone mineral density (BMD) was measured with dual energy X-ray absorptiometry (DXA). Mean BMD of the RA men was lower than in a reference population. Two variables correlated consistently with BMD namely erosive disease and treatment with sulphasalazine. Patients without erosive disease had normal BMD in average, a plausible explanation is that both osteoporosis and erosions in RA is caused by the bone resorbing cells, the osteoclasts, which are activated by proinflammatory cytokines. Sulphasalazine seemed to protect from bone resorption, a finding we intend to investigate further. Women had more severe disease compared with men in the aspects of disease activity and physical disability but not regarding radiographic damage.

The differences between genders largely depended on the worse outcome in women older than 50 years. The more severe disease in older women seemed to be related to menopause state rather than age per se. In women smoking lowered the age at disease onset, but actually only in RF negative women, whereas in men presence of RF and reported relatives with RA were correlated with lower age at disease onset. Also, some of the accepted prognostic factors differed in their ability to predict disease outcome after 2 years disease in men compared with women. Thus initial-markers of inflammation correlated stronger with outcome (DAS28, HAQ score and radiographic score) in women than in men, whereas RF correlated with radiographic outcome in both gender but with DAS28 only in men at the 2 years follow up.

In conclusion, men with RA have aberrations in testosterone, DHEAS, estrone and estradiol but only estradiol concentrations consistently correlate with disease activity. The high estradiol levels in men with high inflammatory RA may be beneficial for bone health. Hormonal factors seem to influence RA incidence and severity.

List of scientific papers

I. Tengstrand B, Carlstrom K, Hafstrom I (2002). Bioavailable testosterone in men with rheumatoid arthritis-high frequency of hypogonadism. Rheumatology. 41(3): 285-9.
https://pubmed.ncbi.nlm.nih.gov/11934965

II. Tengstrand B, Hafstrom I (2002). Bone mineral density in men with rheumatoid arthritis is associated with erosive disease and sulfasalazine treatment but not with sex hormones. J Rheumatol. 29(11): 2299-305.
https://pubmed.ncbi.nlm.nih.gov/12415584

III. Tengstrand B, Carlstrom K, Fellander-Tsai L, Hafstrom I (2003). Anormal levels of serum DHEAS, estrone and estradiol in men with rheumatoid arthritis: high correlation between serum estradiol and current degree of inflammation. Journal of Rheumatology. [Accepted]

IV. Tengstrand B, Ahlmén M, Hafstrom I (2003). The influence of gender on rheumatoid arthritis: a prospective study of onset and outcome after 2 years. [Submitted]