Objective To investigate the predictive values of Kwon's score,Chumbler's score,Age,Atrial Fibrillation,Dysphagia,Sex,Stroke Severity (A2DS2) score,the Preventive ANtibacterial THERapy in Acute Ischemic Stroke (PANTHERIS) score,Acute Ischemic Stroke-Associated Pneumonia Score (AIS-APS),and prestroke Independence,Sex,Age,NIHSS (ISAN) score for stroke-associated pneumonia (SAP) in patients with acute ischemic stroke.Methods The patients with acute ischemic stroke were enrolled retrospectively.They were grouped according to whether to be complicated with SAP or not.The demography and baseline characteristics were compared between the SAP group and the non-SAP group.Multivariate logistic regression analysis was used to identify the independent risk factors for SAP.Receiver operating characteristic (ROC) curves were used to compare the predictive values of the 6 kinds of scores for SAP.Results A total of 1 427 patients with acute ischemic stroke were enrolled.Three hundred ninety-five patients (27.7％) complicated with SAP within 7 d after onset.There were significant differences in age,gender,past history (pneumonia,atrial fibrillation,smoking),laboratory tests (white blood cell count ＞11 × 109/L,baseline blood glucose ≥ 11.1 mmol/L),Oxfordshire Community Stroke Project (OCSP) classification,falling at the time of onset,dysphagia,mechanical ventilation and the modifiel Rankin Scale (mRS) score before onset,baseline Glasgow Coma Scale (GCS) score,baseline National Institutes of Health Stroke Scale (NIHSS) score and 6 scores between the SAP group and the non-SAP group (all P ＜ 0.05).Multivariate logistic regression analysis showed that age (odds ratio [OR] 1.034,95％ confidence interval [CI] 1.019-1.049;P=0.001),white cell count ＞ 11 × 109/L (OR 4.386,95％CI 2.763-6.905;P=0.001),baseline blood glucose ≥ 11.1 mmol/L (OR 1.933,95 ％ CI 1.305-2.864;P =0.001),dysphagia (OR 7.839,95％ CI 4.892-12.563;P =0.001),baseline NIHSS (OR 1.120,95％ CI 1.077-1.165;P =0.001),and baseline GCS score (OR 1.132,95％ CI 1.019-1.257;P =0.021) were the independent risk factors for SAP.The areas under the ROC curves of SAP predicted by the Chumbler's,AIS-APS,A2DS2,ISAN,Kwon's and PANTHERIS scores were 0.830 (95％ CI 0.805-0.855),0.827 (95％ CI 0.802-0.852),0.818 (95％ CI 0.792-0.845),0.788 (95％ CI 0.762-0.814),0.774 (95％CI 0.774-0.803),and 0.727 (95％ CI 0.695-0.758),respectively.There were no significant differences in the area under ROC curves of Chumbler's,A2DS2 and AIS-APS scores between the pairwise comparisons.There were significant differences in the area under ROC curves of the Chumbler's,A2DS2,AIS-APS and ISAN scores between the pairwise comparisons (AIS-APS compared with ISAN:P =0.001;the rests P ＜ 0.001).Conclusions The accuracies of predicting SAP with the Chumbler's,AIS-APS and A2DS2 scores are superior to the ISAN,Known's and PANTHERIS scores,and have higher clinical application value.