PROGNOSTIC FACTORS IN MESANGIAL IGA GLOMERULONEPHRITIS - AN EXTENSIVE STUDY WITH UNIVARIATE AND MULTIVARIATE ANALYSES
PROGNOSTIC FACTORS IN MESANGIAL IGA GLOMERULONEPHRITIS - AN EXTENSIVE STUDY WITH UNIVARIATE AND MULTIVARIATE ANALYSES
复制标题
DOI：
10.1016/s0272-6386(12)80284-8
复制
发表时间：
1991-07-01
影响因子：
13.2
通讯作者：
BERTHOUX, F
中科院分区：
文献类型：
--
作者：
ALAMARTINE, E;SABATIER, JC;BERTHOUX, F
为了明确特发性伊加肾小球肾炎（IgA GN）慢性肾功能衰竭的危险因素，我们对282例IgA GN患者进行了双因素研究，采用标准单变量统计方法和考克斯多变量回归模型。随访1 ~ 36年，平均8年，18%（50/282）的患者发生慢性肾功能衰竭（CRF），其中18例（6%）为终末期肾病（ESRD）。CRF患者与非CRF患者的单因素比较显示多个危险因素：主要是动脉高血压、大量蛋白尿、肾病综合征、高水平的血清伊加、HLA-1335抗原的存在以及光镜下大多数病理病变的强度。肾功能正常（血清肌酐< 135 μmol/L [1.5 mg/dL]）的精算生存率在10年时为84%，在20年时为64%。多变量研究仅分离出对生存率有显著影响的四个危险因素。这些是蛋白尿量，首次肾活检的总体光学评分，初始高血压的存在和HLA-B35抗原的存在。根据这些结果，可以计算个体患者的肾脏存活概率，并定义高风险亚组。我们的数据证实了多变量统计分析在寻找IgA肾病患者CRF的危险因素方面比单变量统计分析更有用。
To clarify the risk factors for chronic renal failure in idiopathic IgA glomerulonephritis (IgA-GN), we performed a dual study on 282 patients using both standard univariate statistical methods and the multivariate regression model of Cox. During a follow-up ranging from 1 to 36 years, with a mean of 8 years, 18% of the patients (50/282) had gone into chronic renal failure (CRF), with 18 of them in end-stage renal disease (ESRD) (6%). The univariate comparison of patients with CRF versus patients without CRF showed multiple risk factors: mainly arterial hypertension, an amount of proteinuria, nephrotic syndrome, a high level of serum IgA, presence of HLA-1335 antigen, and the intensity of most pathological lesions on light microscopy. The actuarial survival rate for a normal renal function (serum creatinine < 135 μmol/L [1.5 mg/dL]) was 84% at 10 years and 64% at 20 years. The multivariate study allowed the isolation of only four risk factors with a significant effect on survival rate. These were the amount of proteinuria, the global optical score on first renal biopsy, the presence of an initial hypertension, and the presence of the HLA-B35 antigen. From these results, the probability of renal survival for individual patients may be calculated and a high-risk subgroup defined. Our data confirm the greater usefulness of multivariate over univariate statistical analyses in finding risk factors for CRF in IgA-GN.