Relationship between prognostic nutritional index and post-stroke cognitive impairment

The Prognostic Nutritional Index (PNI) has been described as a useful screening tool for patient prognosis in several diseases. As a potential diagnostic index, it has attracted the interest of many physicians. However, the correlation between the PNI and post-stroke cognitive impairment (PSCI) remains unclear.

A total of 285 patients with acute ischemic stroke were included. PNI was assessed as serum albumin (g/L) + 5× lymphocyte count (10⁹/L) and was dichotomized according to the prespecified cut-off points 48.43 for the high and low groups. PSCI was defined as Mini-Mental State Examination (MMSE) < 27 at the 6-10 months follow-up. Multiple logistic regression and linear regression analyses were performed to examine the association between PNI and cognitive outcomes.

A low PNI was independently associated with PSCI after adjusting for age, sex, education, National Institutes of Health Stroke Scale (NIHSS), deep white matter hyperintensity (DWMH), and stroke history (odds ratio [OR]: 2.158; 95% confidence interval [CI]: 1.205-3.863). The PNI scores were significantly associated with MMSE and attention domain (β = 0.113, p = 0.006; β = 0.109, p = 0.041, respectively). The PNI improved the model's discrimination when added to the model with other clinical risk factors.

A low PNI was independently associated with the occurrence of PSCI and the PNI scores were specifically associated with the scores of global cognition and attention domain. It can be a promising and straightforward screening indicator to identify the person with impaired immune-nutritional status at higher risk of PSCI.