Abstract:

ABSTRACT Objective：To investigate the affecting factors on the therapeutic response and long-term outcome in patients with crescentic ANCA-associated glomerulonephritis. Methodology：Sixty patients with crescentic AAGN patients diagnosed by renal biopsy were enrolled in this study. They were male 25, and female 30cases, with a median age of 54years old. The average levels of serum creatinine (SCr) was 5.0±2.9mg/dl. According to whether the renal replacement therapy (RRT) was needed, the ptients were classified into two groups: the RRT Group（n=30) and non-RRT Group (n=30). With 3 months of treatment period, if the patients in RRT Group get rid of RRT and the Scr of the patients in non-RRT Group declines by 25% or more, the situation can be defined as good response (GR), otherwise, it should be reduced to no response (NR). Multivariate Cox regression analysis is applied to probe into the correlation between clinical as well as pathological parameters and therapeutic response and long-term outcome. Results：Age, BVAS score and SCr level of RRT Group were apparently higher than those of non-RRT Group. The ratio of the circumferential crescents in crescents was higher in RRT Group than that of non-RRT Group (without statistical differences). As for other pathological indexes including total cellular crescents and the ratio of sclerotic glomeruli, there were no marked differences between the above two groups. After 3 months of treatment, for 53.3% patients in RRT Group and 80% patients in non-RRT Group, their therapeutic response is GR; for 46.7% patients in RRT Group and 20% patients in non-RRT, their therapeutic response is NR. According to multivariate Cox regression analysis, the percentage of the circumferential crescents in crescents was the main factor affecting whether the RRT Group can get rid of RRT. When the percentage of crescents was equal or larger than 50%, there was a significant increase in the risk that the patients cannot get rid of RRT；while in the non-RRT Group, the ratio of normal glomeruli of GR patients was obviously higher than that of NR patients. When the ratio of normal glomeruli was higher than 7%, the therapeutic response was more likely to be GR. Among 60 cases, during a follow-up 1~170 months with the median time of 19 months, 6 patients (10%) were dead and 18 (30%) developed to ESRD. From the multivariate Cox analysis, the need of RRT and massive proteinuria was accounted for the independent risk factors for long-term outcome. Conclusion：The ratio of circumferential crescents in crescents and the ratio of normal glomeruli account for an important factor affecting the treatment response to RRT and non-RRT in patients with crescentic ANCA-associated glomerulonephritis, respectively. The need of RRT and massive proteinuria are main factors for long-term outcome. This indicates that physicians should take into account the combination of clinic and renal pathological parameters to formulate a reasonable therapy plan and evaluate prognosis.

Key words: Antineutrophil cytoplasmic antibody, crescentic nephritis, treatment;prognosis