Relationship of Geriatric Nutritional Risk Index (GNRI) with length of hospitalization and mortality rate in elderly patients

Abstract

The proportion of elderly population is growing faster than any other age group. Malnutrition is
a widespread problem in elderly and has been recognized as the most common cause of mortality
and morbidity. Mini Nutritional Assessment (MNA) is recommended to detect malnutrition risk
among geriatrics. However, it is just appropriate for geriatric who stay at home. For hospitalization
patients, Nutritional Risk Index (NRI) that using albumin and weight is recommended. However,
the weight data is often not possible obtained in geriatrics. Therefore, Geriatric Nutritional Risk
Index (GNRI) using ideal body weight is recommended. The aim of this study was to evaluate
relationship of GNRI with length of hospitalization and mortality rate in geriatric patients. This
was an observational study with a prospective cohort design conducted in Dr. Sardjito General
Hospital, Yogyakarta from January to February 2012. Patients who met the inclusion and exclusion
criteria were recruited. Laboratoty and clinical examinations as well as GNRI scoring were then
performed. Patients were grouped into two groups i.e. patients with GNRI score < 82 and e”
82. The patients were then monitored during hospitalization until they were discharged due to
die or recovered. The length of stay and patients died were then recorded. The results showed
that the length of stay of patients with GNRI score < 82 (14.32 ± 8.20 days) was significantly
longer than those with GNRI score 82 (9.31 ± 6.15 days) (p = 0.006). Moreover, the mortality
rate of patients with GNRI score < 82 (42.1%) was significantly higher than those with GNRI
score 82 (2.8%) (p = 0.000). Kaplan-Meier survival analysis showed survival rate of patients
with GNRI score <82 rapidly decreased when compared with those with GNRI score 82. In
conclusion, there is negative correlation between GNRI with length of hospitalization and mortality
rate in elderly.