Comparison of three PET methods to assess peritoneal membrane transport

The peritoneal equilibration test (PET) is the most widespread method for assessing water and solute transport across the peritoneal membrane. This study compared three methods: traditional PET (t-PET), mini-PET, and modified PET (mod-PET). Non-diabetic adults (n=21) who had been on peritoneal dialysis (PD) for at least three months underwent t-PET (glucose 2.5%-4 h), mini-PET (glucose 3.86%-1 h), and mod-PET (glucose 3.86%-4 h) to determine dialysate-to-plasma concentration ratio (D/P) for creatinine and dialysate-to-baseline dialysate concentration ratio (D/D0) for glucose. Agreement between methods regarding D/P creatinine and D/D0 glucose was assessed using analysis of variance (ANOVA), Pearson's correlation coefficient, and Bland-Altman analysis. D/P creatinine differed between t-PET and mini-PET (P<0.001) and between mod-PET and mini-PET (P<0.01) but not between t-PET and mod-PET (P=0.746). The correlation of D/P creatinine with t-PET vs mod-PET was significant (r=0.387, P=0.009) but not that of t-PET vs mini-PET (r=0.088, P=0.241). Estimated bias was -0.029 (P=0.201) between t-PET and mod-PET, and 0.206 (P<0.001) between t-PET and mini-PET. D/D0 glucose differed between t-PET and mod-PET (P=0.003) and between mod-PET and mini-PET (P=0.002) but not between t-PET and mini-PET (P=0.885). The correlations of D/D0 glucose in t-PET vs mod-PET (r=-0.017, P=0.421) or t-PET vs mini-PET (r=0.152, P=0.609) were not significant. Estimated bias was 0.122 (P=0.026) between t-PET and mod-PET, and 0.122 (P=0.026) between t-PET and mini-PET. The significant correlation of D/P creatinine between t-PET and mod-PET suggested that the latter is a good alternative to t-PET. There was no such correlation between t-PET and mini-PET.