Protein catabolic rate in patients on continuous peritoneal dialysis. A multivariate predictive model

Area of Science:

• Nephrology
• Clinical Nutrition
• Biostatistics

Background:

• Protein catabolic rate (PCR) and PCR normalized to standard weight (PCRN) are crucial nutritional markers for patients undergoing continuous peritoneal dialysis.
• Predicting these markers aids in optimizing patient management and nutritional support.

Purpose of the Study:

• To investigate whether urea clearance parameters predict PCR and PCRN in patients on continuous peritoneal dialysis using multivariate analysis.
• To develop and validate predictive models for low PCR and low PCRN.

Main Methods:

• Stepwise logistic regression was employed on a derivation set of 143 urea kinetic studies (92 patients).
• Models were validated using an independent set of 189 urea kinetic studies (102 patients) via receiver operating characteristic (ROC) curve analysis.
• Urea nitrogen excretion normalized to standard weight (UNEN) was added as a variable in subsequent analyses.

Main Results:

• Low serum urea, high creatinine, low urine/dialysate volumes, and low body surface area predicted low PCR.
• Low serum urea, male gender, high BMI, and low urea fractional clearance (KT/V) predicted low PCRN.
• Low UNEN, male gender, and obesity predicted low PCRN; low UNEN, male gender, low urine volume, low drain volume, and high albumin predicted low KT/V.

Conclusions:

• PCR and PCRN are predictable using models integrating azotemic indices, body size/composition, and urea clearance measures.
• Small body size and lean composition influence PCR and PCRN inversely.
• UNEN predicts both PCRN and KT/V, suggesting a potential mathematical link between them.