In this study, we wanted to evaluate the use of kidney biopsies for estimation of N(glom) and V(glom) in both healthy and chronically diseased kidneys. Danish Landrace pigs with mean weight of 29 kg (range: 25-35 kg) were either subjected to unilateral ureteral obstruction (UUO) or non-obstruction (healthy). N(glom) and V(glom) was estimated by design-based methods using biopsies, N(glom)(biopsy) and V(glom)(biopsy). From each kidney, six biopsies were withdrawn at six topographically different sites. All estimates were done following stereological principles and reference methods estimated number with the physical fractionator, N(glom)(PF), and volume with test point system, V(glom)(TPS). N(glom)(PF) was 1.30 +/- 0.28 x 10(6) for UUO kidneys and 1.59 +/- 0.28 x 10(6) for healthy kidneys. N(glom)(biopsy) was 1.18 +/- 0.28 x 10(6) (p > 0.05) for UUO and 1.33 +/- 0.24 x 10(6) (p = 0.04) for healthy kidneys. When UUO and healthy kidneys were grouped, N(glom)(PF) was 1.46 +/- 0.31 x 10(6), and N(glom)(biopsy) was 1.28 +/- 0.25 x 10(6) (p > 0.05). V(glom)(TPS) was 1,079 +/- 126 mm(3) for UUO and 1,707 +/- 263 mm(3) for healthy kidneys. V(glom)(biopsy) was 1,048 +/- 291 mm(3) for UUO (p > 0.05) and 1,373 +/- 393 mm(3) for healthy kidneys (p > 0.05). When UUO and healthy kidneys were grouped, V(glom)(TPS) was 1,180 +/- 229 mm(3) and V(glom)(biopsy) 1,129 +/- 334 mm(3) (p > 0.05). Biopsy sites were tested for any systematic differences between site- and mean values, and no significant difference was found (p > 0.05). This study showed that biopsies can be used for estimating N(glom) and V(glom) by design-based methods, but more precise determination of biopsy volume is needed.