Objective: Overweight and hyperlipidemia, the two established risk factors for acute ischemic stroke, are paradoxically associated with favorable outcomes. The paradox may be resolved by the concept of protein energy wasting (PEW), in which total cholesterol level and body mass index are used as nutritional indexes for predicting outcomes of chronic kidney disease. Methods: Among 12 271 people with acute ischemic stroke and chronic kidney disease, 2086 were defined as being at risk of PEW—with a body mass index <22 kg/m2 plus either a serum albumin level <38 g/L or a total cholesterol level <4.14 mmol/L (160 mg/dL) without the use of lipid-lowering drugs—and all the others were a control group. The hazards of PEW for mortality and functional outcomes were evaluated using propensity score matching and multivariate Cox regression analysis. Results: Based on the propensity score, 2081 PEW participants were matched to the same number of non-PEW control participants. PEW was associated with a higher mortality risk at 3 mo (adjusted hazard ratio, 1.19; 95% confidence interval [CI], 1.02–1.42) and 1 y (adjusted hazard ratio, 1.33; 95% CI1.13–1.52). PEW was also associated with poor functional outcomes (modified Rankin Scale score >2) at 1 mo (adjusted odds ratio, 1.32; 95% CI, 1.08–1.61) and 3 mo (adjusted odds ratio, 1.27; 95% CI, 1.03–1.56). Conclusions: According to the PEW-based assessment system, a modest decrease in body mass index and total cholesterol levels suggests malnutrition and is associated with adverse outcomes of acute ischemic stroke.

Original languageEnglish
Article number111431
Publication statusPublished - Jan 2022


  • Cholesterol
  • Chronic kidney disease
  • Ischemic stroke
  • Protein energy wasting

ASJC Scopus subject areas

  • Endocrinology, Diabetes and Metabolism
  • Nutrition and Dietetics


Dive into the research topics of 'Protein energy wasting–based nutritional assessment predicts outcomes of acute ischemic stroke and solves the epidemiologic paradox'. Together they form a unique fingerprint.

Cite this