Background Skin disease affects >80% of lupus patients and has been linked to lupus nephritis (LN) flares. We recently found that skin inflammation caused by ultraviolet light leads to increased expression of renal inflammatory and injury markers as well as proteinuria, mediated by neutrophils, which migrated from the inflamed skin to the kidneys. These data prompted the hypothesis that presence of concurrent skin disease in lupus patients leads to worse nephritis mediated by neutrophils.

Methods Lupus nephritis biopsies (n=43) were retrieved from the Dartmouth Hitchcock Medical Center tissue bank and grouped: (i) LN with concurrent skin rash (malar, subacute, or discoid n =15) and (ii) LN without any rash (n=6) at the time of LN diagnosis (class IV). Patients with inconclusive or incomplete skin findings were excluded. The study included only female patients. Clinical and laboratory data (absolute neutrophil (ANC) and lymphocyte (ALC) counts, serum creatinine, glomerular filtration rate (GFR), autoantibody titers, and complement) were collected by chart review. Histologic and immunofluorescence data were extracted from biopsy reports. Student’s t-test was performed to detect statistical significance between patient groups.

Results Presence of skin rash at the time of LN flare was indicative of more active disease, reflected by higher ANA, anti-dsDNA IgG, and low C3 and C4 levels. LN patients with concurrent skin disease had a higher neutrophil count (p < 0.05) and neutrophil-lymphocyte ratio (p< 0.01), but not ALC, compared to LN patients without a concurrent skin rash. High neutrophil counts in the presence of skin disease associated with lower GFR (< 60; p< 0.01). This association was not seen in the absence of concurrent skin disease and rash alone did not predict GFR. Analysis of kidney immunofluorescence revealed that concurrent skin disease at the time of LN flare associated with greater IgA deposition. Increased renal IgA levels associated with higher neutrophil but not lymphocyte counts (p < 0.01). This was specific to IgA as no associations with IgG or IgM deposition were detected.

Conclusions Our study provides novel findings that suggest neutrophils may be the pathogenic link between skin inflammation and LN: (i) higher neutrophil counts are found in the presence of a skin rash at the time of LN flare, (ii) neutrophilia but not lymphocytosis in the presence of a skin rash associates with worse kidney function, and (iii) high neutrophils in the presence of skin rash associate with increased renal IgA. These data provide a novel model of IgA-driven activation of neutrophils leading to kidney injury initiated by skin inflammation.