Subjective global assessment: alternative nutrition-assessment technique for liver-transplant candidates.

The interpretation of most objective nutrition assessment parameters is affected by liver disease. A subjective global assessment (SGA) tool adapted for assessment of adult liver-transplant candidates was evaluated for interobserver reproducibility. Two dietitians assessed 20 liver-transplant candidates in regard to 25 SGA parameters based on patient history, physical appearance, and existing clinical conditions. Fifteen percent of the patients were found to be well nourished, 70% moderately malnourished, and 15% severely malnourished. Raters agreed on the nutrition status of the patients 80% of the time. Twelve individual parameters showed fair to good interrater reliability (kappa > or = 0.500). Muscle wasting and fat depletion were determined to be the strongest predictors of the final SGA rating (kappa = 0.737, p < 0.0001, kappa = 0.632, p < 0.0001, respectively). The assessment tool itself was determined to be reliable by the Cronbach coefficient alpha-test (0.707). Although this assessment method has limitations, the findings of this study suggest that SGA as an alternative test for assessing the nutrition status of adult liver-transplant candidates has an overall fair to good interobserver reproducibility rate.