Association between renal insufficiency and malnutrition in older adults: Results from the NHANES III

Background. The extent to which relevant confounding variables influence th e recognized association between renal insufficiency and malnutrition is no t known. This study examined whether renal insufficiency was associated wit h malnutrition, independent of relevant demographic, social, and medical co nditions in noninstitutionalized adults 60 years of age and older. Methods. Participants (5248) in the United States Third National Health and Nutrition Examination Survey (NHANES III, 1988 to 1994), a cross-sectional study, were examined in a multivariate logistic regression model. Particip ants were stratified into three groups of glomerular filtration rate (GFR) by serum creatinine. Dietary and nutritional factors were estimated from 24 -hour dietary recall, biochemistry measurements, anthropometry, and bioelec trical impedance. Participants were malnourished if they demonstrated at le ast three of the following five criteria: (1) serum albumin less than or eq ual to 37 g/L, (2) male weight less than or equal to 63.9 kg, female weight less than or equal to 51.8 kg, (3) serum cholesterol <4.1 mmol/L, (4) ener gy intake <15 kcal/kg/day, and (5) protein intake <0.5 g/kg/day. Results. A GFR <30 mL/min/1.73 m(2) was present in 2.3% of men and 2.6% of women; these participants demonstrated low energy and protein intake and hi gher serum markers of inflammation. Thirty-one percent of individuals with malnutrition demonstrated a GFR <60 mL/min/1.73 m(2). In multivariate analy sis, a GFR <30 mL/min/1.73 m2 was independently associated with malnutritio n [odds ratio 3.6 (2.0 to 6.6)] after adjustment for relevant demographic, social and medical conditions. Conclusions. It is probable that renal insufficiency is an important indepe ndent risk factor for malnutrition in older adults. Malnutrition should be considered, prevented, and treated as possible in persons with clinically i mportant renal insufficiency. These results should be confirmed in a prospe ctive longitudinal cohort study.