Clinical Characteristics And Pathogenic Genetic Evolution Analysis Of SFTS Patients In A Hospital In Jinan

Severe fever with thrombocytopenia syndrome(SFTS),first discovered in China in 2009,is an emerging hemorrhagic fever,which caused by severe fever with thrombocytopenia syndrome virus(SFTSV).Most SFTS patients initially present with non-specific symptoms,such as fever,decreased platelet and white blood cell counts.Severely ill patients will have symptoms such as hemorrhage and respiratory failure,and the disease progresses rapidly.Deaths can occur within two weeks of the onset.In China,the SFTS case fatality rate is about 16%,and it is higher in South Korea and Japan,at 22%and 27%,respectively.However,no specific antiviral drugs and vaccines have been developed to treat and prevent SFTS.Intensive monitoring and conservative treatment are still the main prevention and treatment measures for SFTS.Therefore,at this stage,by analyzing the clinical data of patients to explore the factors affecting the patient’s disease outcome,early identification of groups with a higher risk of disease progression and appropriate interventions are beneficial to improve the cure rate of critically ill patients and to allocate medical resources more rationally.In addition,intragenic recombination can cause SFTSV antigenic transformation,and the clinical manifestations and mortality of patients infected with different genotypes of SFTSV strains are different,but few studies have explored the relationship between genetic diversity and clinical phenotypes in SFTSV.Therefore,this study collected the serum of suspected SFTS patients in a hospital in Jinan from June to December 2020,and confirmed the diagnosis by laboratory tests.The clinical data of confirmed cases during hospitalization were collected,and they were divided into severe and mild groups on the basis of the clinical data.The differences in clinical and laboratory indicators between the groups were analyzed,and a model for predicting severe risk of patients was established to provide guidance for the formulation of clinical treatment plans for SFTS cases.Moreover,this study obtained SFTSV whole genome sequence through highthroughput sequencing.Then the genetic and evolutionary differences of SFTSV between groups was analyzed,and the genetic evolution of SFTSV strains in Shandong Province were explored,which provid further views into the relationship between SFTSV genetic diversity and disease progression and the variation in the epidemic process of SFTSV.Objective1.To analyze the differences in clinical and laboratory indicators(the first test results after admission)between severe and mild SFTS patients,and establish a model for predicting severe risk of patients.2.To analyze the genetic and evolutionary differences of SFTSV between mild and severe groups,and to explore the genetic evolution of SFTSV strains in Shandong Province.Methods1.Collected the serum of suspected SFTS patients in a hospital in Jinan from June to December 2020,and used nested PCR to confirm the diagnosis.Confirmed cases were included in the modeling group.Then,the clinical and laboratory index data of the confirmed cases during hospitalization were collected.Based on whether the patient has been admitted to an intensive care unit(ICU)and meets one of the following criteria:1)acute lung injury or acute respiratory distress syndrome;2)heart failure;3)acute renal failure;4)encephalitis;5)shock 6)sepsis;7)disseminated intravascular coagulation,divided into mild and severe groups.The differences of 64 variables including demographic characteristics,clinical symptoms,blood routine,five items of blood coagulation and serum biochemistry were compared between the two groups.The confirmed cases of SFTS in this hospital from June to December 2019 were included in the validation group.2.The collected clinical data were screened by Least absolute shrinkage and selection operator(LASSO)regression,and the screened variables were incorporated into Logistic regression to build a predictive model.The Hosmer-Lemeshow goodness-of-fit test was used to evaluate the model fit.The receiver operating characteristic curve(ROC)was used to evaluate the predictive value of the model.Bootstrapping was used for internal validation.3.Inoculated the sera of patients with positive SFTS RNA into monolayer Vero cells,and blindly passed three generations to isolate SFTSV.Nested PCR was used to identify the successfully isolated virus strains.SFTSV universal primers were designed.Then PCR amplification was performed,and the whole genome sequence was obtained by highthroughput sequencing.The SFTSV sequences of Shandong Province as of December 31,2020 were obtained from GenBank,including 64 S segments,51 M segments,and 51 L segments.4.MEGA X constructed phylogenetic trees using maximum likelihood method and and then performed genotyping analysis.The amino acid sequences of SFTSV strains RdRp,GP,NP,and NSs were compared by Metadata-driven Comparative Analysis Tools(meta-CATS)on the ViPR website,and Bioedit was used to perform nucleotide homology analysis.Through Datamonkey Website server,SLAC was used to analyze the overall selection pressure and single site selection pressure of SFTSV RdRp,GP,NP,NSs nucleotide sequences,meanwhile used FUBAR,MEME to detect the genetic pressure of a single site.When the results of the three methods were all significant differences,it indicated that the site was positively selected by selection pressure.Results1.From June to December 2020,a total of 142 serum samples of suspected SFTS patients were collected in a hospital in Jinan.After testing,115 samples were positive for SFTSV RNA.A total of 115 confirmed cases were determined in the end.The modeling group finally included 108 SFTS patients(7 patients with missing data),83 patients in the mild group and 25 in the severe group.The ratio of males in the severe group was higher than that of females(P<0.05).In the severe group,the ratio of chronic diseases,the occurrence of body temperature over 39℃,cough,expectoration,headache or dizziness,neurological symptoms and the median maximum body temperature were significantly higher than those in the mild group(P<0.05).The laboratory indicators of severe and mild patients were analyzed,and it was found that creatine kinase,activated partial thromboplastin time,thrombin time,D-dimer,aspartate aminotransferase(AST),aspartate aminotransferase/alanine aminotransferase(AST/ALT)ratio,alkaline phosphatase,leucine aminopeptidase,glutamate dehydrogenase,adenosine deaminase,urea,serum creatinine(SCR),cystatin C,anion gap,α-hydroxybutyrate dehydrogenase and lactate dehydrogenase levels in the severe group were significantly higher than those in the mild group(P<0.05).The levels of fibrinogen,platelet count(PLT),platelet specific volume(PCT),mean platelet volume(MPV),and prealbumin in the severe group were lower than those in the mild group(P<0.05).2.During the fever period,the white blood cell count(WBC),monocyte count(MONO),red blood cell count,lymphocyte count(LY),and neutrophil count(NEUT),mean corpuscular hemoglobin(MCH),PLT,MPV,PCT,PDW levels were compared between the severe and mild disease groups in the modeling group,and there was no significant difference(P>0.05).During the multiple organ damage stage,compared with the mild group,the severe group had lower LY,PLT and PCT(P<0.05).During the recovery period,NEUT and PDW in the severe group were higher than those in the mild group(P<0.05);while the levels of LY,MONO,PLT and PCT were lower than those in the mild group(P<0.05).3.The variables screened by LASSO regression were neurological symptoms and AST/ALT ratio,and they were included in Logistic regression analysis.The results showed that neurological symptoms(OR,13.568;95%CI,3.485-52.813;P<0.001)and AST/ALT ratio(OR,1.817;95%CI,1.215-2.717;P=0.004)was an independent risk factor for the development of severe disease.The mathematical equation of the risk prediction model:M=1/((1+e^(-4.334+2.608×neurological symptoms+0.597×(AST/ALT)))).The HosmerLemeshow goodness-of-fit test was performed on the models constructed in the modeling group,and the results showed that the fitting effect was good(P=0.829.The area under ROC curve(AUC)value of the model constructed by the modeling group was 0.880(95%CI:0.803-0.958);the sensitivity was 88%;the specificity was 76%.The AUC value of the model constructed by internal validation was 0.881(95%CI;0.804-0.958);the sensitivity was 88%;the specificity was 76%.The AUC value of the model constructed by external validation was 0.873(95%CI:0.775-0.972);the sensitivity was 100%;the specificity was 66%.4.115 serum samples from patients with positive nucleic acid test were inoculated into monolayer Vero cells,and passed blindly for 3 generations.Finally,SFTSV was successfully isolated from 94 samples.The complete genome sequences of 18 SFTSV strains were obtained using PCR and highthroughput sequencing,of which 10 virus strains were isolated from the serum of severe patients,and 8 were isolated from mild patients.5.No reassortant strains were found in the 18 newly obtained SFTSV strains.The virus strains obtained in the mild group included C1 subtype(2 strains)and C3 subtype(6 strains);the severe group included C1 subtype(3 strains),C3 subtypes Subtype(6 strains)and C4 subtype(1 strain).What’more,the type distribution of each fragment of the virus strains obtained in the two groups was similar.The sequences obtained in GenBank showed that there were four different genotypes in Shandong Province,including C1-C4 subtypes,of which C3 subtype was the main one.The amino acid sequences of the four proteins of the virus strains were compared between the mild and severe groups,and no statistically significant differences were found(P>0.05).The amino acid sequence of the newly obtained virus strain was compared with that of the SFTSV strain in Shandong Province in GenBank,and 1,4,and 1 site with statistically significant differences were found in the amino acid sequences of RdRp,GP,and NSs,respectively(P<0.05).6.Combining the newly obtained sequences in this study with the SFTSV sequences in Shandong Province in GenBank,it was found that the nucleotide homology of each fragment of SFTSV strains in Shandong Province was relatively high;RdRp,GP,NP,and NSs were all negatively selected.The 37th locus of the GP gene was a positive selection site.Conclusions1.Neurological symptoms and AST/ALT ratio were independent risk factors for the development of severe disease in SFTS patients.The robustness of the SFTS severity prediction model constructed based on these two indicators was high and can be used as a reference for formulating clinical treatment plans.2.There was no significant difference in the type distribution and amino acid sequence of SFTSV strains between the mild and severe groups,and the genetic diversity of SFTSV isolates was not found to significantly affect the disease progression of the patients.3.SFTSV strains in Shandong Province were closely related,and various subtypes of SFTSV strains were prevalent in Shandong Province,including C1～C4 subtypes,among which C3 subtype was the main one.The amino acid sequences of the newly obtained virus strain and the SFTSV RdRp,GP,NP,and NSs obtained in GenBank were compared,and there were 6 amino acid sites with significant differences.