CT测量肾脏深度优化Gates法测定肾小球滤过率

Renal depth measured by CT optimize the glomerular filtration rate using Gates method

目的探讨CT测量肾脏深度校正和优化的采集及后处理方法在Gates法测定肾小球滤过率(GFR)中的应用价值。方法回顾性分析2018年1月至2019年11月在华中科技大学同济医学院附属协和医院行肾动态显像Gates法测定GFR的患者157例[男102例,女55例,年龄(51.4±14.5)岁],包括正常肾区组(肾脏位置和形态正常,且排除肾积水、肾占位及腹膜后包块等影响肾脏深度因素的成人患者)118例和非正常肾区组39例(移植肾19例,马蹄肾11例,异位肾9例)。对正常肾区组分别采用CT测量、传统Tonnesen公式、李乾公式(简称Li法)计算肾脏深度,并得到相应GFR;对非正常肾区组,分别采用优化的采集和后处理方法、传统的后处理方法、慢性肾脏病流行病学协作组(CKD-EPI)推荐的基于酶法测定血清肌酐(Scr)估算法得到的GFR(分别用GFR优、GFR传和eGFR表示)。采用单因素方差分析和最小显著差异t检验比较正常肾区组Tonnesen公式和Li法所得的肾脏深度及相应GFR与CT实测法的差异、非正常肾区组GFR优和GFR传与eGFR的差异;采用Pearson相关分析和Bland-Altman分析评价不同方法计算GFR的相关性和一致性。结果正常肾区组,CT实测左、右肾脏深度分别为(7.40±1.43)和(7.51±1.37)cm。Tonnesen公式低估了肾脏深度[左肾(6.03±0.82)cm,右肾(6.06±0.84)cm;F值:64.145和68.567,均P<0.01],且肾脏深度偏差随CT实测深度的增加而增大(r值:0.847和0.834,均P<0.01)。Tonnesen公式相应的总肾GFR为(56.93±28.42)ml·min^-1·1.73 m^-2,与CT实测法[(73.43±36.56)ml·min^-1·1.73 m^-2]的差异有统计学意义(F=9.423,P<0.01);Li法估算的左、右肾脏深度分别为(7.55±1.03)和(7.52±0.98)cm,相应的总肾GFR为(73.65±34.50)ml·min^-1·1.73 m^-2,与CT实测法的差异均无统计学意义(均P>0.05);Li法所得GFR与CT实测法相关性(r=0.901,P<0.01)和一致性更好。非正常肾区组,GFR优、GFR传和eGFR分别为(63.11±27.40)、(48.40±25.45)和(59.89±32.24)ml·min^-1·1.73 m^-2,其中GFR传与GFR优的差异有统计学意义(F=2.870,P=0.025);GFR优与eGFR相关性(r=0.941,P<0.01)和一致性更好。结论Tonnesen公式低估了肾脏深度且存在局限性,利用CT图像测量肾脏深度,进行深度校正,可以提高Gates法测定GFR的准确性。对于移植肾、马蹄肾、异位肾及腹膜后包块等特殊情况,优化采集方案及后处理方法对于获得准确的GFR非常重要。

Objective To explore the application value of CT measurement of renal depth correction,optimized acquisition and post-processing in the measurement of renal glomerular filtration rate(GFR)by Gates renal dynamic imaging.Methods From January 2018 to November 2019,157 patients(102 males,55 females,age(51.4±14.5)years)including 118 in normal renal area group(adults with normal renal position and morphology,and excluding hydronephrosis,renal occupation,retroperitoneal mass and other factors affecting renal depth)and 39 in abnormal renal area group(19 of transplanted kidney,11 of horseshoe kidney and 9 of ectopic kidney),were retrospectively enrolled in Union Hospital,Tongji Medical College,Huazhong University of Science and Technology.The GFR was measured by renal dynamic imaging Gates method.For the normal renal area group,the renal depth was calculated by CT method,the traditional Tonnesen formula or the Li Qian formula.For the abnormal renal area group,the GFR was measured by optimized acquisition and post-processing method(GFR optimization),the traditional post-processing method(GFR tradition),or Chronic Kidney Disease Epidemiology Collaboration(CKD-EPI)formula method(eGFR).The differences of the renal depth and corresponding GFR obtained by different methods were analyzed using one-way analysis of variance and the least significant difference(LSD)t test.The correlation was analyzed by Pearson correlation analysis,and the consistency was analyzed by Bland-Altman analysis.Results In the normal renal area group,the left and right renal depth measured by CT were(7.40±1.43)and(7.51±1.37)cm.Tonnesen formula underestimated renal depth(left kidney:(6.03±0.82)cm,right kidney:(6.06±0.84)cm;F values:64.145 and 68.567,both P<0.01),and the deviation increased with the increase of CT measured depth(r values:0.847 and 0.834,both P<0.01).The GFR measured by Tonnesen formula was(56.93±28.42)ml·min^-1·1.73 m^-2,and the difference was statistically significant compared with CT method((73.43±36.56)ml·min^-1·1.73 m^-2;F=9.423,P<0.01).The renal left and right depth measured by Li Qian formula were(7.55±1.03)and(7.52±0.98)cm,and the total GFR was(73.65±34.50)ml·min^-1·1.73 m^-2 with no differences compared with CT method(all P>0.05).The GFR obtained by Li Qian formula had better correlation(r=0.901,P<0.01)and consistency with CT method.In the abnormal renal area group,GFR optimization,GFR tradition and eGFR was(63.11±27.40),(48.40±25.45)and(59.89±32.24)ml·min^-1·1.73 m^-2,respectively,and the difference between GFR tradition and GFR optimization was statistically significant(F=2.870,P=0.025).GFR optimization had better correlation(r=0.941,P<0.01)and consistency with eGFR.Conclusions Tonnesen formula underestimates the renal depth.Using CT to measure renal depth and perform depth correction can improve the accuracy of Gates method for GFR determination.For the special cases of transplanted kidney,horseshoe kidney,ectopic kidney and retroperitoneal mass,it is important to optimize acquisition scheme and post-processing method to obtain accurate GFR.