MALNUTRITION PREDICTS LONG-TERM MORTALITY IN HOSPITALIZED GASTROENTEROLOGICAL PATIENTS

 

Introduction:

Malnutrition is a common problem in gastroenterological (GI) diseases. Subjective Global Assessment (SGA), Nutritional Risk Screening (NRS) and Bioelectrical impedance analysis (BIA) are well established methods to clinically evaluate a nutritional (NT) status in patients (pts).

Objectives:

Assessment of the predictive potential of NT status parameters as characterized by SGA, NRS, phase angle (PhA) and serum surrogate parameters, resp. for overall long-term mortality in gastroenterological pts.

Methods:

NT status was assessed in 652 GI pts in a cross-sectional study by SGA and NRS scores. In addition, BMI, serum albumin were determined and BIA performed. These pts were followed for a mean period of 60 months (range: 57 – 89 months). Overall outcome analysis was performed using the Kaplan-Meier Method. For multivariate analysis of independent prognostic factors of overall survival the Cox proportional-hazard model was applied.

Results:

462 (70.9%) pts were found as well-nourish by using SGA screening and 482 (73.9%) pts by using NRS screening, resp. By screening an increased risk of or manifest malnutrition (NRS > 3, SGA scores B/C) was found in 190 pts (29.1%, SGA) and in 170 pts (26.1%, NRS). Malnourished pts had significantly lower BMI, serum albumin level, median phase angle as compared with the well-nourish group. As relevant clinical endpoint overall survival was analysed in two subgroups separately: pts with benign (Be) or malignant (M) GI diseases. Overall 1- and 5- year survival rates (YSR) were significantly shorter in malnourished pts (SGA B+C; NRS < 3 resp.) in both groups (p < 0.001, Be+M). An increased relative risk of death was associated with PhA < 4.7 ° (Be: 5-YSR 61.8% vs. 41.8%; M: 5-YSR 49.3% vs. 5%) and serum albumin < 35 mg/dl (Be: 5-YRS 79.1% vs. 54.2%). By multivariate analysis SGA (Be: HR 3.015, p < 0.001; M: HR 2.924, p < 0.001), NRS (Be: HR 2.056, p < 0.011; M: HR 3.14 p < 0.001), PhA cut off 4.7 ° (Be: HR 2.431, p < 0.011; M: HR 2.032, p < 0.004) and albumin were found to be independent risk factors of long-term mortality.

Conclusion:

Malnutrition is highly prevalent in hospitalized GI-pts and is associated with distinct clinical diagnoses. In the present analysis we demonstrate that malnutrition not only predicts short-term but also long-term survival in GI-pts. NT assessment by SGA, NRS, serum albumin level and BIA could also predict overall outcome.

Disclosure of Interest:

None declared.