Background Environmental contamination with mercury is linked to increased incidence of autoimmune diseases. Exposure to low-dose non-organic mercury has been reported to exacerbate murine SLE.

Objectives To study the serum mercury(Hg) levels in patients with SLE and their associated clinical risk factors

Methods Consecutive patients who fulfilled >=4 ACRcriteria for SLEin a 2-month period were recruited from our out-patient clinics. An equal number of age and sex matched healthy control subjects who attended the Hong Kong Red Cross for blood donation was also recruited for comparison. Blood was taken from subjects in the morning for the assay of Hg level (atomic absorption spectrophotometry by a DMA80 direct Hg analyzer, Milestone, Italy) and other markers of disease activity (anti-dsDNA, complement C3 and C4) (for SLE patients). Demographic and clinical data of SLE patients were collected. Disease activity score was assessed by the SLE disease activity index (SLEDAI) and organ damage was assessed by the SLICC/SLE damage index (SDI). Bivariate correlation between Hg level and various clinical and serological markers was studied by Pearson’s correlation test. A linear regression model was established to study the clinical risk factors associated with higher serum Hg levels in the recruited patients.

Results 246Chinese SLEpatients(93% women) were studied. The mean age was 40.6±12.5years and SLE duration was 7.7±7.0years. The mean SLEDAI and SDI score was 4.9±5.8 and 0.97±1.5, respectively. 73 (30%) patients had clinically active SLE, defined as a SELENA-SLEDAI score of >=6. 107 (43%) patients had organ damage (SDI>=1). The mean serum total mercury level of the SLE patients studied was 1.34±0.69 ng/mL (NR <3.85), which was significantly higher than that of 246 healthy subjects (N=246) (0.72±0.34 ng/mL; p<0.001). Only 2(0.8%) SLE patients had Hg level >3.85ng/mL. Patients with clinically active SLE (SLEDAI>=6) had significantly higher Hg level than those with SLEDAI<6 (1.52±0.56 vs 1.26±0.74 ng/mL; p=0.004). In addition, the levels of mercury in patients with clinically and serologically inactive disease (SLEDAI =0) were significantly higher than those with controls (1.15±0.56 vs 0.72±0.34 ng/mL; p<0.001) adjusted for age and sex (by ANCOVA). Bivariate correlation study revealed that younger age (r -0.24; p<0.001), female sex (r0.19; p=0.003), shorter SLE duration (r-0.16; p=0.01), anti-dsDNA titer (r0.22; 0.001), SLEDAI score (r0.26; <0.001) and physicians’ global assessment of disease activity (r0.18; p=0.008) were significantly associated with higher Hg level. SDI score (r-0.06; p=0.33), C3 (r-0.06; p=0.36), C4 (r-0.07; p=0.28), photosensitivity, renal insufficiency (estimated creatinine clearance <50ml/min) and other clinical manifestations were not significantly associated with mercury levels. In a linear regression model, the SLEDAI score (beta 0.16; p=0.02) and female sex (beta 0.17; p=0.008) were independently associated with higher Hg level, after adjustment for age, SLE duration and renal insufficiency.

Conclusions In this cross-sectional study, the total serum Hg level was significantly elevated in patients with both active and inactive SLE than matched controls. Hg level correlated independently with clinical disease activity and the female sex in a multivariate model, suggesting mercury may be an environmental trigger for SLE.

Disclosure of Interest None Declared