Is There an Association between Glomerular Density and Birth Weight in Healthy Humans?

Dear Sir,It has been estimated that the total number of nephrons in the normal human population ranges from 300,000 to 1,200,000 or more per kidney [1, 2, 3]. Brenner et al. [2]and Brenner and Chertow [3]have hypothesized that individuals with fewer nephrons at birth are at higher risk of developing essential hypertension in later life and at enhanced risk of expression of renal disease. They further suggested that a congenital deficit of nephrons might be caused by intrauterine growth retardation, as it occurs in infants with low birth weight. Some studies have already confirmed this hypothesis by demonstrating that a low birth weight is associated with a reduction in the total number of nephrons [4], with more frequent development of arterial hypertension [5], and with a poor prognosis of some renal diseases [6, 7, 8].The significance of nephron number is, therefore, now well appreciated. However, little is known about the glomerular density in the normal human kidney. It is not clear to what extent the glomerular density varies in healthy individuals and which factors influence the glomerular density in the normal human kidney. We, therefore, analyzed the glomerular density in autopsy kidney samples of presumably healthy individuals who died accidentally. We further wished to find out whether there is an association between birth weight and glomerular density in the normal human kidney.Our study included kidney autopsy samples of 20 presumably healthy individuals (7 women and 13 men, aged 18–25 years, mean age 21.5 years) who died accidentally and had no macroscopic or histologic autopsy evidence of renal or cardiovascular diseases, diabetes, or other major pathologic conditions. Data on birth weight were obtained from the birth records.All renal autopsy specimens were fixed in buffered formalin, embedded in paraffin, and sectioned at 3–4 µm. The slices were stained with haematoxylin-eosin and periodic acid-Schiff. Morphometric analysis was performed by using the image analysis system IBAS-1000 (Kontron). We measured the cortical area in each autopsy sample and counted all glomeruli within this area. The glomerular density was expressed as number of glomeruli per square millimetre.The measured areas of the renal cortex ranged from 14.44 to 94.55 mm2 (54.4 ± 21 mm2; mean ± SD), and the number of glomeruli within these areas ranged from 44 to 245 (156 ± 58.4). The birth weights ranged from 1,800 to 4,200 g (3,225 ± 710 g). The correlation between glomerular density and birth weights is shown in figure 1. A statistically significant correlation was found between glomerular density and birth weights (r = 0.46; p < 0.05).The results of our preliminary study suggest that the glomerular density in the kidneys of healthy humans varies and that it could be related to the birth weight, similar to the total number of nephrons. The total number of nephrons (glomeruli) can be estimated by stereological techniques [1, 4]which are time-consuming and can be only used if the whole kidney is available. In contrast, determination of the glomerular density is simple and rapid and can be performed using archival histologic material obtained at autopsies or biopsies.In contrast to the total number of nephrons (glomeruli), little is known about the glomerular density in the normal kidney in humans or experimental animals. Leroy et al. [9]noticed reduced numbers of glomeruli per optical field in low-birth-weight infants as compared with those with appropriate birth weight. In the present study, we analyzed kidney autopsy samples of 20 presumably healthy individuals and found 2.1–3.4 glomeruli per square millimetre of renal cortical tissue. A statistically significant correlation was found between glomerular density and birth weights.We conclude that the glomerular density is a more practical tool for studying renal structure and function than the total number of glomeruli. Being related to birth weight, it might also prove to be a prognostic factor in renal disease. However, this hypothesis should be confirmed in a larger number of cases.