Using Nutrition Risk Scores to Predict Hospital Length of Stay in Mild Acute Pancreatitis: A Prospective Cohort Study
Kittikundecha T, MD1, Lakananurak N, MD2, Rerknimitr R, MD3
Affiliation : 1 Department of Medicine, Faculty of Medicine, Chulalongkorn University and King Chulalongkorn Memorial Hospital, The Thai Red Cross Society, Bangkok, Thailand. 2 Division of Clinical Nutrition, Department of Medicine, Faculty of Medicine, Chulalongkorn University and King Chulalongkorn Memorial Hospital, The Thai Red Cross Society, Bangkok, Thailand. 3 Division of Gastroenterology, Department of Medicine, Faculty of Medicine, Chulalongkorn University and King Chulalongkorn Memorial Hospital, The Thai Red Cross Society, Bangkok, Thailand
Background : Malnutrition increases both morbidity and mortality in patients with acute pancreatitis (AP). However, few studies
have investigated degree of malnutrition and their impact on patients’ outcomes.
Objective : This study is to investigate the incidence of malnutrition in hospitalized patients with AP and the effect of malnutrition
on length of stay (LOS).
Materials and Methods : Data from all patients who were admitted with AP at King Chulalongkorn Memorial Hospital from April
2016 to July 2017 were collected. The degree of malnutrition was evaluated using Nutritional Risk Screening (NRS) 2002 and
Subjective global assessment (SGA). The effects of malnutrition on hospital LOS were calculated by Spearman’s correlation (rs).
Results : A total of 32 patients with AP were recruited. Twenty-two (68.8%), eight (21.9%) and two (9.4%) patients were classified
as having mild, moderately severe and severe AP by revised Atlanta criteria 2012. Fourteen (43.8%) patients were categorized as
at risk of malnutrition by NRS 2002 (score >3) and 10 (31.2%) patients were diagnosed with malnutrition by SGA (SGA class B and
C). NRS 2002 score, but not SGA, significantly predicted LOS (rs = 0.5, p = 0.003), which mean LOS (SD) of NRS score 1 to 4 were 3.5
(1.2), 8.6 (8), 8.2 (5.9) and 11.6 (5.7) days, respectively.
Conclusion : Malnutrition was common in patients admitted with AP. As for validated nutrition tools, NRS 2002 had better efficacy
than SGA to predict LOS in AP patients.
Keywords : Acute pancreatitis, Malnutrition, Length of stay, Nutrition risk screening, Subjective global assessment
