The Research Of Association Between Peripheral Blood NPAR And The Severity Of Autoimmune Encephalitis

Objective:To analyze the clinical characteristics and prognosis of autoimmune encephalitis(AE)related to anti-neuronal surface antigen antibody,to investigate the meaning of neutrophil percentage-to-albumin ratio(NPAR)in assessing the severity of AE,and to contrast the clinical features of two frequent antibodies encephalitis,in order to offer clinical reference for the diagnosis,treatment and prognosis of AE.Method:The data of patients with AE diagnosed for the first time in Qilu Hospital of Shandong University from January 2018 to August 2022 were collected,including general information,clinical features,AE-related antibodies,laboratory tests,auxiliary examination,treatment and prognosis.(1)According to the modified Rankin scale(mRS),the mRS score≤3 was divided into the non-severe group,and the mRS score>3 was divided into the severe group.The clinical data with statistically significant variances and excluding collinearity were analyzed by multivariate Logistic regression analysis applying forward stepwise approach to construct the independently influential factors of severe AE nomogram.The Bootstrap approach was utilized for internal validation,and the receiver operating characteristic curve(ROC)was plotted,the area under the curve(AUC)was computed,and the Hosmer-Lemeshow goodness-of-fit test was utilized for reflecting the discrimination and calibration of the model.(2)The most common AE in the group were anti-NMDAR antibody encephalitis and anti-LGI1 antibody encephalitis,which were divided into two groups according to the type of antibody to explore their clinical difference.Results:A group of 126 AE patients were enrolled in this study,including 47(37.3%)anti-NMDAR,44(30.9%)anti-LGI1,14(11.1%)anti-GABABR,5(4.0%)anti-CASPR2,4(3.2%)anti-mGluR5,1(0.8%)anti-DPPX and 11(8.7%)multiple antibodies.This study included 69(54.8%)cases in the non-severe group,with anti-LGI1 antibody being the most common in 27(39.1%)cases,57(45.2%)cases in the severe group,with anti-NMDAR antibody being the most common in 26(45.6%)cases.Univariate analysis indicated that NPAR,NLR,ESR,ALT,AST,urea nitrogen,creatinine,HCY,mental and behavioral disorders,sleep disorders,antibody titer of cerebrospinal fluid,and tumor were the influencing factors of severe AE(P<0.05).Multivariate Logistic regression analysis demonstrated that NPAR,urea nitrogen and mental behavior abnormalities were independent influencing factors of severe AE.The risk of severe AE increased with the increase of NPAR and urea nitrogen levels(OR=12.682,95%CI 2.916～55.164,P=0.001;OR=1.378,95%CI 1.018～1.866,P=0.038).Mental and behavioral disorders were associated with a higher risk of severe AE(OR=8.167,95%CI 3.105～21.484,P<0.001).The ROC analysis indicated that the area under the curve of the prediction model was 0.839(95%CI:0.768～0.910,P<0.001),NPAR was 0.734(95%CI:0.647～0.821,P<0.001),urea nitrogen was 0.604(95%CI:0.768～0.910,P<0.001)and mental and behavioral abnormalities was 0.717(95%CI:0.624～0.810,P<0.001).The best cut-off value of NPAR calculated by ROC was 1.795,the sensitivity was 52.6%and the specificity was 87.0%.Hosmer-Lemeshow goodness of fit test indicated that χ2=5.855,P=0.663>0.05,demonstrating that the model had good calibration and high prediction efficiency.To analyse of clinical data of anti-NMDAR and anti-LGI1 antibodies encephalitis,the two groups were significantly different in gender,age,prodromal symptoms,seizures,mental and behavioral abnormalities,memory loss,neutrophil count,neutrophil percentage,HDL-C,blood sodium,carcinoembryonic antigen,NLR,intracranial pressure,cerebrospinal fluid white blood cells,cerebrospinal fluid lactic acid,cerebrospinal fluid biochemistry(glucose,chloride,protein),serum antibody titer,cerebrospinal fluid antibody titer,cerebrospinal fluid oligoclonal band,combined tumor,mRS Score at admission and mRS Score at the last follow-up.Conclusion:1.The common antibodies in AE were anti-NMDAR and anti-LGI1.The incidence of severe AE with anti-NMDAR antibody was higher than that with anti-LGI1 antibody.2.NPAR,urea nitrogen and mental and behavioral abnormalities were independent influencing factors of severe AE.3.NPAR was an independent predictor of the severity of AE,and the risk of severe AE increased with the increase of NPAR level.The risk of severe AE was higher when NPAR≥1.795,while that of NPAR<1.795 was lower.4.The AUC and 95%CI of the prediction model was 0.839(95%CI:0.768-0.910),the NPAR was 0.734(95%CI:0.647-0.821),urea nitrogen was 0.604(95%CI:0.505-0.703)and mental behavior abnormalities was 0.717(95%CI:0.624-0.810),indicating that the diagnostic prediction model of severe AE improved the predictive effectiveness of NPAR,urea nitrogen and mental behavior abnormalities alone.