HDCN Article Review/Hyperlink

Malhotra D, Tzamaloukas AH, Murata GH, Fox L, Goldman RS, Avasthi PS

Serum albumin in continuous peritoneal dialysis: Its predictors and relationship to urea clearance

Kidney Int (Jul) 50:243-249 1996

This excellent study has two components. Step wise logistic regression is used to identify the predictors of serum albumin, in 92 CAPD patients. The findings are then successfully validated in a further 102 patients. Predictors are found to be, in order of significance, high peritoneal transport status on PET, older age, presence of diabetes and a high daily dialysate drain volume. It is notable that neither Kt/V, creatinine clearance or protein catabolic rate were predictive and, although uncorrected PCR was higher in those with a higher serum albumin, it did not come out as positive in the multiple regression analysis. These findings support very similar results obtained by our group and published in 1993 [JASN 3:1501]

In the second part of the study, the authors ask whether prospective increases in peritoneal dialytic dose can increase serum albumin. Thirty-seven patients are studied, an average of eight months after an increase in mean weekly Kt/V from 1.4 to 2.1. The normalised PCR had increased from 0.74 to 0.96 but the serum albumin was unchanged. Within the 37 patients, however, there was a subgroup of 15 in whom the albumin did increase. This subgroup were notable in that they were younger than the group as a whole and that they also had a greater increase in their serum urea after the increase in Kt/V. They also showed a somewhat higher nPCR both before and after the intervention. Comment:

The implication of this study is that serum albumin in PD patients is a complex index and its main determinants are not dialytic dose nor PCR. Rather, factors that are not easily modified such as peritoneal transport status, age, and co-morbidity are more influential. Prospective increases in dialytic dose should be still be considered in the treatment of malnutrition in PD patients but a response in the serum albumin should not be reliably be expected. (Peter G. Blake, M.D., Victoria Hospital, London, Ontario)