A critical approach to nutritional assessment in critically ill patients

Background and aims: Nutritional assessment enhances quality of nutritional care, however, its practice bemuses professionals. This prospective study aimed to identify a feasibie/informative nutritional parameter in intensive care. Methods: 44 patients (APACHE II: 23.8+/-10.1), age 58.4+/-18.6 years, were evaluated at admission: clinical data, height, weight, body mass index (BMI), tricep skinfold thickness, mid-arm circumference (MAC), mid-arm muscle circumference (MAMC), albumin, total protein and lymphocyte count. Anthropometric parameters' performance was evaluated isolated or assembled according to Blackburn and McWhirter criteria. Results: Oedema increased %IW and BMI (P < 0.01); muscle depletion was frequent and agreed with MAC or MAMC ranked by both criteria, P = 0.02. %IW and BMI overestimated well-nourished/overweight patients, whilst arm anthropometry, mostly MAC/MAMC, shifted towards +/-50% malnutrition. Patients were not equally ranked by both criteria; McWhirter's by using percentiles clarified the distribution and showed agreement between MAC and MAMC, P = 0.007, unlike Blackburn's. Mortality was higher in patients with MAC < 5th percentile, P = 0.003; MAC < 15th percentile was able to predict mortality and major complications. In invasive ventilated patients, severe muscle depletion was associated with mortality, P = 0.05, Conclusion: In intensive care most nutritional assessment methods are useless; MAC is simple, feasible and if classified by percentiles may prove functional with prognostic value. (C) 2002, Elsevier Science Ltd. All rights reserved.