The current study was undertaken to investigate the relation between serum C-reactive protein (CRP) concentrations and parameters of renal function in dogs with naturally occurring renal disease. Dogs were assigned to groups according to plasma creatinine concentration, urinary protein-to-creatinine ratio (UP/UC), and exogenous plasma creatinine clearance (P-Cl(Cr)) rates. Group A (healthy control dogs; n = 8): non-azotemic (plasma creatinine <125 µmol/l) and nonproteinuric (UP/UC <0.2), with P-Cl(Cr) rates >90 ml/min/m(2); group B (n = 11): non-azotemic, nonproteinuric dogs with reduced P-Cl(Cr) rates (50-89 ml/min/m(2)); group C (n = 7): azotemic, borderline proteinuric dogs (P-Cl(Cr) rates: 22-67 ml/min/m(2)); and group D (n = 6): uremic, proteinuric dogs (not tested for P-Cl(Cr)). The serum CRP concentrations were measured via commercial enzyme-linked immunosorbent assay. The CRP concentrations in the clinically healthy dogs (group A) ranged from 2.09 mg/l to 8.60 mg/l (median: 3.21 mg/l). In comparison with dogs of group A, median CRP concentrations were significantly (P < 0.01) elevated in dogs of group B (17.6 mg/l, range: 17.0-19.2 mg/l), group C (24.8 mg/l, range: 18.0-32.5 mg/l), and group D (59.7 mg/l, range: 17.7-123 mg/l). Serum CRP was significantly related to P-Cl(Cr) (r = -0.83; P < 0.001), plasma creatinine (r = 0.81; P < 0.001), UP/UC (r = 0.70; P < 0.001), and leukocytes (r = 0.49; P < 0.01). The significant relations between serum CRP concentrations and biochemical parameters of kidney function in plasma and urine suggest that a stimulation of the acute phase response is implicated in the pathogenesis of canine renal disease.