BACKGROUND: Factors predictive of renal outcome were investigated in 219 cases of biopsy-proven Henoch-Schönlein purpura nephritis (HSPN); 83 children and 136 adults enrolled in a national study were followed up for up to 27 years (median, 4.5 years). METHODS: The criterion for defining disease progression was time elapsed until doubling of baseline creatinine level and until dialysis therapy. Age, sex, data at onset (renal function, proteinuria, hematuria, hypertension, and crescents), and data during follow-up (proteinuria and therapy) were tested as covariates. RESULTS: Multivariate Cox regression analysis indicated the following parameters as independent prognostic predictors: age (adults versus children, relative risk, 3.57; 95% confidence interval, 1.18 to 10.79; P = 0.024 for creatinine level doubling; relative risk, 14.89; 95% confidence interval, 1.72 to 129.07; P = 0.014 for dialysis therapy), sex (females versus males, relative risk, 5.71; 95% confidence interval, 1.67 to 19.55; P = 0.006 for creatinine level doubling; relative risk, 26.03; 95% confidence interval, 2.64 to 256.73; P = 0.005 for dialysis therapy), and mean proteinuria during follow-up (for each 1 g/d of protein increase, relative risk, 1.77; 95% confidence interval, 1.35 to 2.32; P < 0.001 for creatinine level doubling; relative risk, 1.73; 95% confidence interval, 1.18 to 2.52; P = 0.005 for dialysis therapy). Information for mean proteinuria levels during follow-up increased the sensitivity at logistic regression to 62.5%, with dialysis therapy as the end point. No data detected at diagnosis, including renal function impairment, proteinuria, hypertension, and crescentic nephritis (involving > 50% of glomeruli in only 2.6%), were significantly related to functional decline at multivariate Cox. CONCLUSION: This analysis indicates that, even more than when decreased renal function, severe proteinuria, hypertension, or crescents are present at onset, the risk for progression of HSPN (greater in adults and females) was associated with increasing mean proteinuria levels during follow-up.
BACKGROUND: Factors predictive of renal outcome were investigated in 219 cases of biopsy-proven Henoch-Schönlein purpura nephritis (HSPN); 83 children and 136 adults enrolled in a national study were followed up for up to 27 years (median, 4.5 years). METHODS: The criterion for defining disease progression was time elapsed until doubling of baseline creatinine level and until dialysis therapy. Age, sex, data at onset (renal function, proteinuria, hematuria, hypertension, and crescents), and data during follow-up (proteinuria and therapy) were tested as covariates. RESULTS: Multivariate Cox regression analysis indicated the following parameters as independent prognostic predictors: age (adults versus children, relative risk, 3.57; 95% confidence interval, 1.18 to 10.79; P = 0.024 for creatinine level doubling; relative risk, 14.89; 95% confidence interval, 1.72 to 129.07; P = 0.014 for dialysis therapy), sex (females versus males, relative risk, 5.71; 95% confidence interval, 1.67 to 19.55; P = 0.006 for creatinine level doubling; relative risk, 26.03; 95% confidence interval, 2.64 to 256.73; P = 0.005 for dialysis therapy), and mean proteinuria during follow-up (for each 1 g/d of protein increase, relative risk, 1.77; 95% confidence interval, 1.35 to 2.32; P < 0.001 for creatinine level doubling; relative risk, 1.73; 95% confidence interval, 1.18 to 2.52; P = 0.005 for dialysis therapy). Information for mean proteinuria levels during follow-up increased the sensitivity at logistic regression to 62.5%, with dialysis therapy as the end point. No data detected at diagnosis, including renal function impairment, proteinuria, hypertension, and crescentic nephritis (involving > 50% of glomeruli in only 2.6%), were significantly related to functional decline at multivariate Cox. CONCLUSION: This analysis indicates that, even more than when decreased renal function, severe proteinuria, hypertension, or crescents are present at onset, the risk for progression of HSPN (greater in adults and females) was associated with increasing mean proteinuria levels during follow-up.
Authors: Outi Jauhola; Jaana Ronkainen; Olli Koskimies; Marja Ala-Houhala; Pekka Arikoski; Tuula Hölttä; Timo Jahnukainen; Jukka Rajantie; Timo Örmälä; Matti Nuutinen Journal: Pediatr Nephrol Date: 2012-02-05 Impact factor: 3.714