Nutritional treatment is associated with longer survival in patients with pancreatic disease and concomitant risk of malnutrition.

BACKGROUND: Pancreatic diseases involve complex nutritional challenges. Despite this, conflicting evidence exists regarding the clinical relevance of detecting the risk of malnutrition and implementing systematic nutrition support for these patients. Thus, our aims were to investigate whether screening for malnutrition risk and initiating nutrition support are predictive of mortality for hospitalized patients with pancreatic diseases. DESIGN: From 2008 to 2018, 34 prevalence surveys of nutrition were conducted at Haukeland University Hospital (HUH), Norway. Risk of malnutrition was defined by a score of ≥3 in Nutritional Risk Screening 2002 (NRS 2002). Primary outcomes included overall, one-year, and one-month mortality, and were compared according to malnutrition risk and nutrition support for adult patients with ICD-10 codes of K85: acute pancreatitis, K86: other diseases of pancreas, and C25: malignant neoplasm of pancreas. Length of hospital stay (LOS) was included as a secondary outcome. RESULTS: Of the 283 patients investigated, risk of malnutrition was present in 61.5%. Risk of malnutrition was associated with higher overall mortality (Hazard Ratio (HR) = 1.67, 95% confidence interval (CI): 1.2-2.4, P = 0.003) and one-year mortality (HR = 1.89, 95% CI: 1.2-2.9, P = 0.004) compared to patients not at risk. Not receiving nutrition support for at-risk patients was associated with higher overall mortality (HR = 1.60, 95% CI: 1.1-2.4, P = 0.019) and one-year mortality (HR = 1.64, 95% CI: 1.04-2.6, P = 0.034) compared to patients at risk who received nutrition support. Patients at risk of malnutrition had increased LOS (20.5 nights vs 15.2 nights, P = 0.044) compared to patients not at risk of malnutrition. CONCLUSION: This study of hospitalized patients with pancreatic disease suggests that risk of malnutrition may be associated with higher mortality rates, whereas nutrition support may decrease mortality rates. CLINICAL TRIAL REGISTRY: Not registered.