Usefulness of the prognostic inflammatory and nutritional index (PINI) in hospitalized elderly patients.

The prognostic inflammatory and nutritional index (PINI) is a simple scoring system of overall health which aggregates two blood markers of inflammatory (C-reactive protein and alpha(1)-acid glycoprotein) and of nutritional (albumin and transthyretin) states. This study was undertaken with a view to evaluate, in comparison to currently used predictive approaches, the potential usefulness of PINI to forecast hospital mortality and outcome of patients hospitalized in an acute geriatric unit. 1,066 elderly patients, aged 82.7 +/- 6.6 years and fulfilling inclusion criteria, were enrolled in the study. Logistic regression analysis and calculation of relative risk (RR) were carried out for epidemiological data with a cut-off value of 25 for PINI. Immediate mortality (7.9%) of admissions) was predicted by PINI > or = 25 (RR = 4.34). Only 387 patients (36.3%) could rejoin their residence location (home or family). A sizeable proportion of acute patients (55.8%) failed to recover and/or developed diseased states requiring chronic care management. Incapacity to return home was predicted by PINI > or = 25 (RR = 2.04). Hypoalbuminaemia < or = 30 g/L was not found a predictor of mortality but was associated with total disability (RR = 9.08). The optimal PINI cut-off value to predict mortality was calculated at 8.8 using the ROC analytic approach. We conclude that the PINI formula is helpful to predict both nearest lethality and chronic institutionalization. This scoring system should take a place within the battery of tests used to identify and to follow up acutely ill elderly patients at risk of major complications.