Feasible use of estimated height for predicting outcome by the geriatric nutritional risk index in long-term care resident elderly

Abstract

Background: The Geriatric Nutritional Risk Index (GNRI) is a new index recently introduced for predicting risk of nutritional-related complications in elderly patients. It combines albumin with information about body weight: GNRI = (1.489 x albumin, g/l) + (41.7 x present/ideal body weight), with ideal weight calculated according to the Lorentz formula. Because standing height (SH) is frequently difficult to obtain in older people, in Lorentz equations this parameter has been replaced by estimated height (EH) from knee height. Though, if EH is well accepted as a valid surrogate for SH, the same might not be expected for its use in ideal body weight calculation, with possible consequences in grading nutritional risk correctly.

Objective: The aim of this study was to investigate whether the use of SH rather than EH for the calculation of ideal body weight predicts similar outcomes by GNRI.

Methods: Body weight, SH and EH were obtained in 231 long-term care resident elderly (88 males and 143 females, mean age +/- SD 80.0 +/- 8.4, range 65-97 years). Blood samples were assessed for albumin concentration. Ideal body weight was derived from the Lorentz formula using both SH and EH. According to both ideal weight estimates, nutritional risk was defined by the GNRI score.

Results: The Pearson correlation coefficients were high for both EH (with SH; r = 0.90) and estimates of ideal body weight (r = 0.90) and all were highly significant (p < 0.0001). A statistically significant difference was found between SH and EH (p = 0.0265). Similar and expectable differences in significance have also been observed between ideal body weights (p = 0.0271). However, an accordance of 95.2% has been detected (Kendall's tau test: tau = 0.85, p < 0.0001) in grading nutritional risk by GNRI.

Conclusion: The use of EH for ideal body weight calculation and nutritional risk assessment by GNRI is feasible. Thus, GNRI seems to have been designed in the best way and its use is really attractive, particularly when considering the low-grade participation demanded of the patient in the assessment. This simple and valid assessment tool should be taken into greater consideration.