BACKGROUND: The number of nephrons in humans varies considerably under normal circumstances, and retarded intrauterine growth has been reported to be associated with a significant reduction in nephron number. Low nephron number may be an independent risk factor for the development of hypertension. We therefore decided to evaluate the relationship between body weight at birth and the number and size of nephron units. METHODS: We examined coronal sections of the kidneys of 35 neonates who died within two weeks of birth because of hyaline membrane, infectious complications, brain hemorrhage, or perinatal hypoxia and had no urinary congenital malformations. Nine of them (5 males and 4 females) were between 36 and 37 weeks of gestation, and the rest had 38 or more weeks of gestation. Eighteen neonates weighed less than 2500 g at birth [low birth weight (LBW); 9 females and 9 males], and 17 had weights above this value [normal birth weight (NBW); 8 females and 9 males]. In each section, glomeruli present in four sequential subcapsular microscopic fields, corresponding to 0.6 mm2, were counted; in addition, the area of each of 65 consecutive glomeruli was determined by a computerized measurement system. Glomerular volume was calculated from the glomerular area. Linear regression analysis was used to test the relationship between glomerular number and size and the weight at birth. RESULTS: The number of glomeruli per 0.6 mm2 of renal cortex was 92.9 +/- 4.85 in the LBW and 105.8 +/- 3.91 in NBW (P < 0.0001). Glomerular volume (micro(3) x 10(-3)) was 529.1 +/- 187.63 in the LBW group and 158.0 +/- 49.89 in the NBW group (P < 0.0001). The glomeruli occupied 8.59 +/- 1.38% of the kidney area under examination in the LBW group and 14.3 +/- 2.75% in the NBW group (P < 0.0001). There were significant direct correlations between the weight at birth and the number of glomeruli (r = 0.870, P < 0.0001) and area occupied by glomeruli (r = 0.935, P < 0.0001). There were inverse correlations between the number of glomeruli and the volume of the glomeruli (r = -0.816, P < 0.0001) and the weight at birth and glomerular volume (r = -0.848, P < 0.0001). These findings were independent of sex and race (black vs. white). Essential arterial hypertension existed in 38.9% of the mothers of children with LBW and in 5.9% of the mothers of children with NBW (P < 0.05). Smoking habits existed in 50% of the mothers of LBW children and in 11.8% of the mothers of NBW children (P < 0.05). CONCLUSION: There are strong correlations between glomerular number (direct) and size (inverse) with LBW in this cohort. Endowment with decreased nephron numbers may be a risk factor for hypertension and the rate of progression of renal disease.