Hormonal and Environmental Factors for Systemic Lupus Erythmatosus

Accomplishments: A population-based case-control study of recently diagnosed patients living in eastern North Carolina and South Carolina was conducted: 255 cases and 355 controls frequency matched by age, gender, and state completed the study. Ninety percent of the cases are female and 60% are African-American. Cases were identified through 30 community-based rheumatology and 4 university-based rheumatology practices, public health clinics, and patient support groups. Population-based controls were identified through driver's license records. Almost half (48%) of the patients have been referred by community-based rheumatologists. Data were collected using a standardized interview that includes assessment of previous medical history, family medical history, work and other activities with the potential of exposure to specific materials, and for women, reproductive and menstrual history. Serum was collected to measure specific autoantibodies. White blood cells are stored for additional studies of gene-environment interactions relating to metabolism, immune function, and hormones Preliminary analysis of medical-related risk factors have been completed.
<P>This analysis was limited to conditions that occurred before the age at diagnosis (or reference age for controls) and were adjusted for age, sex, race, and state. The risk of developing SLE increased with a history of medication allergy (OR 3.1, 95% CI 2.1-4.5). The strongest medication-specific association was with allergy to sulfa drugs (OR 2.8, 95% CI 1.5-5.3) . SLE patients were more likely to report a history of herpes zoster (OR 2.5, 95% CI 1.1-5.9). None of the subjects were taking immunosuppressant drugs at the time herpes zoster was diagnosed. There was no association between SLE risk and history of inflammatory conditions (eczema, asthma, hay fever), allergy to foods, animals, or insect stings, or the hepatitis B vaccine. A history of stroke, blood clot, or pulmonary embolism was also more common in SLE patients (OR 5.4, 95% CI 20-15). <P>These results raise questions about the role of specific infectious and allergic conditions in the development of SLE. Preliminary analysis of reproductive and hormonal risk factors have also been completed. Age at natural menopause was younger among cases compared with controls (OR 2.4, 95% CI 1.3, 4.4). This analysis censored observations at surgical menopause or reference (diagnosis) age, so does not reflect an earlier menopause as the consequence of SLE treatment. There was little evidence of an association with use of hormone replacement therapy or oral contraceptives, or with gravidity, parity, or spontaneous abortions. A history of pre-eclampsia was significantly associated with subsequent risk of developing SLE (OR 4.2. 95% CI 2.0-9.0). A decreased risk was seen with breastfeeding when assessed as the number of babies that were breastfed (OR 0.3, 95% CI 0.1-0.8 for 3 or compared to zero) or as total months of breastfeeding, summed across pregnancies (OR 0.4, 95% CI 0.2-0.9 for 52 or more weeks compared to zero). These results suggest that early natural menopause may reflect an underlying autoimmune condition or susceptibility. The increased risk we observed with preeclampsia could be related to microchimerism and an increased maternal-fetal cell circulation in preeclampsia. The mechanism through which lactation would decrease SLE risk is not known. Although estrogen levels are suppressed during periods of frequent lactation, other measures reflecting decreased estrogen exposure in our study were not associated with a decreased risk of SLE.