Renal immunofluorescence and the prediction of renal outcome in patients with proliferative lupus nephritis.

The risk for endstage renal failure in patients with proliferative lupus nephritis (PLN) depends largely on the severity and reversibility of the inflammatory process as determined by light microscopy (LM). As the intrarenal formation of immune complexes is thought to initiate this inflammation, we studied whether renal immunofluorescence microscopy (IFM) provides clinical or prognostic information in addition to LM findings. Clinical data at the time of renal biopsy and during a mean follow-up of 46 months were extracted from the records of 69 SLE patients with proliferative LN (WHO class III/IV). Biopsy specimens were analyzed by LM for AI and CI, while IFM was performed on cryostat sections with the use of antisera against IgG, IgM, IgA, C3, C1q and fibrin. IFM findings were recorded in terms of the localization (glomerular, tubular or vascular) and intensity of fluorescence (score from zero to three). IFM findings were then related to clinical and LM findings and its prognostic value studied by survival analysis. Glomerular immune deposits were present in 99% of patients, tubular deposits in 38% and vascular deposits in 17%. A 'full-house' pattern (all three Ig classes) was found in 67% of biopsies and C3 and C1q deposits in 93% and 74% respectively. Median scores for AI and CI were 6 (1-18) and 3 (0-10); aside from a negative correlation between IgA deposits and CI, we found no other correlation between the amount or type of immune deposits and AI or CI. IgM deposits were associated with high serum levels of anti-dsDNA, while IgG deposits correlated with high ESR and serum creatinin levels. IFM scores were not related to steroid dose at the time of biopsy and neither type of glomerular, tubular or overall renal immune deposits had prognostic value for renal survival. Renal immunofluorescence does not reflect light microscopy findings in patients with PLN and does not contribute prognostic information in patients with PLN. Lupus (2000) 9, 504-510.