Clinical Nutrition Open Science (Apr 2024)

Comparison of nutritional screening and assessment tools for predicting the composite outcome of mortality and complication in cirrhosis

  • Piyanant Chonmaitree,
  • Asawin Sudcharoen,
  • Piyakorn Poonyam,
  • Worawut Roongsangmanoon,
  • Kitsarawut Khuancharee,
  • Nutthawut Laoarphasuwong

Journal volume & issue
Vol. 54
pp. 51 – 62

Abstract

Read online

Summary: Background & Aim: Malnutrition is a common problem among cirrhotic patients. There is currently no gold standard of nutritional assessment tools. The purpose of present study was to compare various nutritional screening and assessment tools for predicting the composite outcome of mortality and complications in cirrhotic patients. Methods: A total of 94 patients with cirrhosis underwent nutritional screening by Royal Free Hospital-Nutritional Prioritizing tool (RFH-NPT), Liver Disease Undernutrition Screening Tool (LDUST), Malnutrition Universal Screening Tool (MUST), Prognostic Nutritional Index (PNI–O), Nutritional Risk Index (NRI), Spanish Society of Parenteral and Enteral Nutrition (SENPE) and Controlling Nutritional Status (CONUT). Nutritional assessment was performed using body mass index (BMI), triceps skinfold (TSF), midarm circumference (MAC), midarm muscle circumference (MAMC), six-minute walk test (6MWT), Subjective Global Assessment (SGA), and Royal Free Hospital Subjective Global Assessment (RFH-SGA). Patients were followed up for 1 year. Result: When compared with other nutrition screening tools, NRI and CONUT had the highest sensitivity (83.3%). NRI also demonstrated the highest negative predictive value (90.7%) and accuracy, reflected by an area under the curve of 0.800, in predicting the composite outcome of mortality and complications of cirrhosis. LDUST and PNI–O showed the highest specificity (89.1%) and positive predictive value (74.1%). In the realm of nutritional assessment, RFH-SGA had the highest sensitivity (73.3%) and negative predictive value (86.4%). BMI demonstrated the highest specificity (98.5%) and positive predictive value (83.3%). SGA showed the highest accuracy, as indicated by an area under the curve of 0.794. Conclusions: NRI and SGA exhibited the highest accuracy in predicting the composite outcome of mortality and complications among cirrhotic patients.

Keywords