Assessment of nutritional status in CAPD patients: serum albumin is not a useful measure.

INTRODUCTION: In CAPD patients serum albumin is frequently used as an index of nutritional status, although it is recognized that hypoalbuminaemia may be caused by many factors. We have further examined the relationship between serum albumin and nutrition.
METHODS: Nutritional status was assessed by biochemistry, anthropometry, mid-arm muscle circumference, muscle strength (hand grip and back), and lean body mass (from anthropometry, creatinine kinetics and bioimpedance) in a group of 76 stable CAPD patients. Correlations between biochemical and nutritional parameters were sought and data were compared between patient groups defined by serum albumin (> or = 37 vs < 37 g/l on two occasions 2 months apart) and separately according to subjective global assessment score (normal nutrition, A vs mild to moderate, B, and severe, C, malnutrition).
RESULTS: In patients with a low SGA score, actual body weight, body mass index, mid-arm muscle circumference, lean body mass, subscapular skinfold thickness, hand grip strength (males and females) and iliac and triceps skinfold thicknesses and back strength (females only) were all significantly less than in patients with a normal SGA score. In contrast, none of these variables differed in either gender when patients were compared according to serum albumin. Serum albumin was correlated with serum creatinine (r = 0.45, P = 0.01), daily urine protein excretion (r = -0.42, P = 0.02) and uncorrected weekly creatinine clearance (r = -0.39) in females, but not with any index of body composition in either gender.
CONCLUSION: Whilst SGA identified a patient group with significantly abnormal body mass, muscle mass and muscle strength, serum albumin did not. Serum albumin is not a useful marker of malnutrition in stable patients on CAPD.