We studied the peritoneal transport properties in 175 stable continuous ambulatory peritoneal dialysis (CAPD) patients seeking a simple and handy assessment of peritoneal permeability to small solutes. Measurement of creatinine in biological fluid was known to suffer from interference by high glucose concentration in the sample. Furthermore, the interference is also affected by the creatinine concentration of the specimen. Peritoneal transport properties were studied by determining the dialysate to plasma ratio of creatinine concentration (D/P) at the fourth hour of the peritoneal equilibration test, and the mass transfer area coefficient of creatinine (MTACCr) or glucose (MTACGlu). The ratio of glucose concentration in peritoneal dialysate effluent (PDE) at 4 and 0 h (G4/G0) was examined and compared with various peritoneal parameters. There were significant logarithmic correlations between D/P or G4/G0 with MTACCr (r=0.96 and 0.79, respectively, p < 0.0001). The correlation between G4/G0 and D/P was linear (r=-0.82, p < 0.0001). A fairly good agreement was present between G4/G0 and D/P by Bland and Altman's method. The bias was -0.93% with 95% confidence interval -23.29% to 21.43% of the measured value. Systematic error was found when D/P or G4/G0 were compared with MTACCr. D/P under estimated MTACCr in the high range. The reverse happened for G4/G0. Net ultrafiltration (NUF) also correlated with MTACCr, D/P and G4/G0 (r=-0.32, p < 0.001; -0.26, p < 0.01; and 0.16, p < 0.05, respectively. In conclusion, the use of G4/G0 as a measure of peritoneal transport in CAPD is an acceptable alternative to D/P. It is highly reproducible and avoids correction of interference when creatinine transport parameters are measured. Because of the logarithmic relations of G4/G0 (or D/P) with MTACCr, the former should not be directly converted to MTACCr. Such a simple measure of peritoneal permeability is, however, most convenient for serial monitoring and can be useful to detect early loss of ultrafiltration or solute clearance.