Abstract
The Geriatric Nutritional Risk Index (GNRI) is used to assess nutritional risk in the elderly population based on the percent ideal body weight (IBW) and albumin (Alb) level, while severe infections are associated with hypoproteinemia and hypoalbuminemia, which can make the treatment of infections challenging. Although GNRI is an effective index to assess disease severity, its association with vancomycin (VCM) -induced renal dysfunction has not yet been clarified. In the present study, classification and regression tree (CART) analysis and receiver operating characteristic (ROC) curve were used to determine the GNRI threshold level that discriminates patients with and without renal dysfunction, with a total of 293 patients who were administered VCM being included. We examined the association between GNRI and the incidence of VCM-induced renal dysfunction to explore ways in which GNRI may be used to determine the target VCM trough levels for individual patients. Based on the threshold level, patients were further divided into high and low GNRI groups, and survival analysis was performed based on trough levels (<20, 20-25, ≥25 μg/mL). Bearing in mind that, due to safety concerns, guidelines do not recommend the trough level of >20 μg/mL, in the high GNRI group (≥68, n = 163), there was no significant difference in the rate of renal dysfunction between those with trough levels of <20 and 20-25 μg/mL (p = 0.66), while in the low GNRI group (<68, n = 130), patients with the trough level of 20-25 μg/mL were at a higher risk of developing renal dysfunction than those with trough level of <20 μg/mL (p < 0.01). While this was in agreement with our low GNRI group, our findings suggest that the trough level may be increased to 25 μg/mL in patients who are in the high GNRI group.
