Nutritional status in peritoneal dialysis. Studies in body composition, lipoprotein metabolism and peritoneal function

Abstract

Nutritional status in peritoneal dialysis. Studies in bodycomposition, lipoprotein metabolism and peritoneal function. Department of Nephrology,Institute of Internal Medicine, Sahlgrenska University Hospital, SE-413 45 Gothenburg,Sweden.The aim of the study was to investigate the nutritional and metabolic consequences and theperitoneal transport conditions of long term peritoneal dialysis. A further purpose was toevaluate methods currently used for measuring delivered dialysis dose and nutritionalstatus.The study was a clinical observational study including almost all patients commencingperitoneal dialysis at the Sahlgrenska University Hospital during the 1990s.Nutritional status was assessed by body composition analyses including total body potassiumas a measure of body cell mass. Total body water was measured by tritium dilution and bodyfat was calculated according to a four-compartment model. At start of dialysis, body cell massand body fat were slightly reduced compared to normal values. During follow-up, there was afurther decline of body cell mass, while energy stores were replenished. Total body waterremained unchanged, which implies an increase in extracellular water.A cross-sectional detailed analysis of the lipoprotein profile showed that the patients onperitoneal dialysis had increased plasma concentrations of cholesterol and triglycerides andincreased levels of ApoB-containing lipoproteins compared to normal subjects. ApoA-I- andApoA-II-containing lipoproteins were reduced and inversely correlated to time on dialysis.Peritoneal function was assessed by the Peritoneal Dialysis Capacity test, based on the three-poremodel of capillary transport. Patients with diabetic nephropathy and nondiabeticsystemic disease had increased peritoneal diffusion capacity already at start of dialysis.Peritoneal protein losses were increased in patients with more severe co-morbidity. Peritonealfunction remained essentially stable during medium long-term follow-up.A comparison was made between estimations of lean body mass from creatinine generationrate and measurements of total body potassium, respectively. A low degree of agreementbetween the two methods was found. Furthermore, the coefficient of variation for repeatedmeasurements of creatinine generation rate was high.The clearance for urea (Kt/Vurea) is normalised to total body water, which is the distributionvolume for urea (V). Prediction equations, based on anthropometric measurements, arefrequently used to estimate V. Measurements of total body water volumes by tritium dilutionwere compared to calculations of total body water according to the Watson formulas.Although mean values of the two methods were similar, the prediction equationsoverestimated the body water in obese patients, and vice versa in lean or overhydratedpatients.These studies have provided further insight in the nutritional status and lipoprotein profile inpatients treated with peritoneal dialysis. Moreover, the limitations of conventional methodsfor nutritional assessment in dialysis patients have been recognized.