Predictors and outcome of renal flares after successful cyclophosphamide treatment for diffuse proliferative lupus glomerulonephritis

Abstract

Objective. To study the incidence, predictors, and outcome of renal flares after successful cyclophosphamide (CYC) treatment for diffuse proliferative glomerulonephritis (DPGN) in patients with systemic lupus erythematosus (SLE). Methods. Between 1988 and 2001, patients with biopsy-proven SLE DPGN who were treated initially with prednisone and CYC were studied. Those who responded to CYC were followed up for the occurrence of renal flares. The cumulative risk, predictors, and outcome of renal flares were evaluated. Results. We studied 189 patients (167 women; and 22 men) with SLE DPGN. All were initially treated with prednisone and CYC (49% orally; 51% by intravenous pulse). At the last dose of CYC, 103 patients (55%) and 52 patients (28%) had achieved complete and partial renal responses, respectively. Azathioprine (AZA) was given as maintenance therapy in 117 patients (75%). After a mean followup of 96.5 months, 59 patients (38%) experienced renal flares (42% nephritic; 58% proteinuric). The median time to relapse was 32 months. The cumulative risk of renal flare was 28% at 36 months and 44% at 60 months. Independent predictors of nephritic flares were persistently low C3 levels after CYC treatment and absence of AZA maintenance therapy. At the last clinic visit, 16 patients (10.3%) had developed doubling of the serum creatinine level (cumulative risk of creatinine doubling 7.4% at 5 years after renal biopsy and 14.3% at 10 years). Ten patients (6.5%) developed end-stage renal disease (ESRD). Renal survival rates at 5 and 10 years were 94.9% and 87.5%, respectively. Increasing histologic chronicity scores, failure to achieve complete response, persistent hypertension after CYC treatment, and nephritic renal flares were unfavorable factors for doubling of the serum creatinine level and for ESRD by univariate analysis. The occurrence of nephritic flares was the only predictor of creatinine doubling by Cox regression analysis. Conclusion. In patients with SLE DPGN, renal flares are common despite initial responses to CYC. Nephritic renal flares are associated with a decline in renal function. Maintenance therapy with AZA reduces, but does not completely prevent, renal flares. More effective maintenance treatment for SLE DPGN after an initial response to CYC should be evaluated.