摘要
目的评价三种肾小球滤过率估算方程在南京地区慢性肾脏病(CKD)人群中的适用性。方法回顾性分析2010年1月至2015年1月在江苏省中医院肾内科住院的168例符合条件患者的临床资料,以99 mTc-DTPA肾动态显像法所测肾小球滤过率(GFR)为参考标准,以r GFR表示;分别用EPI-Scr、EPI-CysC、EPI-scr-cysc方程估算患者GFR,以e GFR表示。分别从偏倚、10%及30%的准确性(P10及P30)、精确度、95%一致性及相关性等方面对方程的适用性进行评估。结果在绝对偏倚方面,三种方程间比较差异均有统计学意义(P<0.01);EPI-Scr的绝对偏倚明显大于EPI-CysC及EPI-scr-cysc,差异均有统计学意义(P<0.016 7),但EPI-CysC及EPI-scr-cysc间比较差异无统计学意义(P>0.016 7)。在偏倚百分比方面,三种方程间比较差异均有统计学意义(P<0.01);EPI-Scr的偏倚百分比明显大于EPI-CysC及EPI-scr-cysc,差异均有统计学意义(P<0.016 7);但EPI-CysC及EPI-scr-cysc间比较差异无统计学意义(P>0.016 7)。一致性限度由小到大分别为EPI-scr-cysc(67.8)、EPI-CysC(68.8)及EPI-Scr(80.6)。与r GFR的Spearman相关系数由大到小分别为EPI-CysC(r=0.774)、EPI-scr-cysc(r=0.769)及EPI-Scr(r=0.559)。P10方面三者间比较差异无统计学意义(P>0.05)。P30方面,三者间比较差异有统计学意义(P<0.01);EPI-Scr的准确性小于EPI-CysC及EPI-scr-cysc,差异均有统计学意义(P<0.016 7);但EPI-CysC及EPI-scr-cysc间比较差异无统计学意义(P>0.016 7)。EPI-scr-cysc及EPI-CysC与EPI-Scr相比,精确度得到了提高,标准误差分别为17.27、17.55、23.39。结论对于南京地区CKD患者来说,EPI-CysC与EPI-scr-cysc方程适用性更佳,但仍需进一步改良并开发更合适的方程以更好服务于临床。
Objective To evaluate the applicability of three calculation equations of glomerular filtration rate in patients with chronic kidney disease(CKD) in Nanjing. Methods The clinical data of 168 eligible patients, who admitted to Jiangsu Province Hospital of TCM from January 2010 to January 2015, were retrospectively analyzed.99 m Tc-DTPA renal dynamic imaging method was used to measure GFR expressed as r GFR. EPI-Scr, EPI-CysC and EPI-scr-cysc were used to estimate GFR expressed as e GFR. The applicability of the equations was evaluated by the bias, 10% and 30% accuracy(P10 and P30), accuracy, 95% consistency and correlation. Results In the absolute bias, the differences between the three equations were statistically significant(P0.01). The absolute bias of EPI-Scr was greater than that of EPI-CysC and EPI-scr-cysc(P0.016 7), but there was no significant difference between EPI-CysC and EPI-scr-cysc(P0.016 7). In the percentage of bias, the differences between the three equations were statistically significant(P0.01). The percentage of bias of EPI-Scr was greater than that of EPI-CysC and EPI-scr-cysc(P0.016 7), but there was no significant difference between EPI-CysC and EPI-scr-cysc(P0.016 7). The 95% confidence interval order that from small to larger was EPI-scr-cysc(67.8), EPI-cysc(68.8) and EPI-Scr(80.6). The order of spearman correlation coefficient of e GFR that from larger to small was EPI-CysC(r=0.774) and EPI-scr-cysc(r=0.769), EPI-Scr(r=0.559).There was no significant difference in P10 among the three groups. However, in P30, EPI-Scr was less accurate than EPI-CysC or EPI-scr-cysc(P0.016 7). There was no significant difference between EPI-CysC and EPI-scr-cysc(P0.016 7). The standard deviation of EPI-CysC and EPI-Scr was 17.55 and 23.39, respectively, which was more accurate than 17.27 of EPI-scr-cysc. Conclusion For CKD patients in Nanjing area, the EPI-CysC and EPI-scr-cysc equations may be better, but further improvements for more suitable equations are still needed.
作者
朱玮玮
倪杰
许嵘
朱晓雷
ZHU Wei-wei NI Jie XU Rong ZHU Xiao-lei.(Department of Nephrology, Jiangsu Province Hospital of TCM, Affiliated Hospital of Nanjing University of TCM, Nanjing 210000, Jiangsu, CHINA)
出处
《海南医学》
CAS
2017年第1期96-98,共3页
Hainan Medical Journal
关键词
南京地区
肾小球滤过率
计算方程
慢性肾脏病
CKD-EPI方程
Nanjing area
Glomerular filtration rate
Calculation equations
Chronic kidney disease(CKD)
CKD-EPI equations