Hinchliffe S A, Sargent P H, Howard C V, Chan Y F, van Velzen D
Department of Fetal and Infant Pathology, University of Liverpool, Royal Liverpool Children's Hospital, United Kingdom.
Lab Invest. 1991 Jun;64(6):777-84.
The disector method, a stereologic procedure unbiased by feature size, shape, or tissue-processing methods, for the estimation of total glomerular number was performed on pairs of human kidneys from 11 normal spontaneous second trimester abortions and stillbirths (15 to 40 weeks gestation). In addition, gestational age-dependent patterns of change in the average volume of the nephron and its cortical and medullary segments were analyzed. Mean glomerular number, plateauing at 36 weeks, increased from 15,000 at 15 weeks to 740,000 at 40 weeks. Average volume of the medullary nephron segment (Henle's loop) increased throughout pregnancy. Average volume of the cortical nephron segment (Tubuli Contorti) decreased from 15 weeks to 25 weeks, then increased after 36 weeks. Fractional volume of the renal cortex decreased from 15 weeks to 40 weeks. Three to 4 hours of microscopic analysis time were required per kidney on routinely processed 5-microns hematoxylin and eosin-stained paraffin sections. Average coefficient of error for number estimation was 8.02%. Average intra- and interobserver reproducibilities were 96.8 and 93.7%, respectively. The demonstrated temporal differences in the development of the cortical and medullary nephron components may result in a dissociation of function, which may explain the increased incidence of fetal hydrops in the second trimester of pregnancy, and which must be taken into account in the treatment of (very) premature infants. Although the number of kidneys included in this study is limited, as they reflect the whole period of antenatal development relevant to neonatal intensive care, the disector method of glomerular number estimation shows significant potential for the analysis and increased understanding of the development of renal function. The method appears to be more sensitive in detecting small and early deviations from normal renal growth and development than previously available parameters e.g., renal weight and (cortical) volume.
采用无偏倚的体视学方法——分割法（该方法不受特征大小、形状或组织处理方法的影响），对11例妊娠中期自然流产和死产（妊娠15至40周）的成对人肾进行总肾小球数量的估算。此外，还分析了肾单位及其皮质和髓质段平均体积随孕周变化的模式。平均肾小球数量在36周时趋于稳定，从15周时的15,000个增加到40周时的740,000个。髓质肾单位段（亨氏袢）的平均体积在整个孕期都在增加。皮质肾单位段（曲管）的平均体积从15周降至25周，然后在36周后增加。肾皮质的体积分数从15周降至40周。对于常规处理的5微米苏木精和伊红染色石蜡切片，每个肾脏需要3至4小时的显微镜分析时间。数量估算的平均误差系数为8.02%。观察者内和观察者间的平均重复性分别为96.8%和93.7%。皮质和髓质肾单位成分发育过程中表现出的时间差异可能导致功能分离，这可能解释了妊娠中期胎儿水肿发生率增加的原因，并且在（极）早产儿的治疗中必须予以考虑。尽管本研究纳入的肾脏数量有限，但由于它们反映了与新生儿重症监护相关的整个产前发育阶段，肾小球数量估算的分割法在分析和增进对肾功能发育的理解方面显示出巨大潜力。该方法在检测与正常肾脏生长发育的微小和早期偏差方面似乎比以前可用的参数（如肾脏重量和（皮质）体积）更敏感。
