Comparison Of Clinical Characteristics Between Severe Fever With Thrombocytopenia Syndrome And Hemorrhagic Fever With Renal Syndrome And Risk Prediction Model Of Differential Diagnosis

Background:Severe fever with thrombocytopenia syndrome(SFTS)is an emerging infectious disease caused by a virus belonging to the Bunyaviridae family,known as severe fever with thrombocytopenia syndrome virus(SFTSV),which is primarily spread by tick bites.Its main clinical manifestations are fever,thrombocytopenia,digestive tract symptoms,and multiple organ dysfunction.In severe cases,patients may experience seizures,coma,shock,and other life-threatening complications.In China,the mortality rate dropped from about 20%in 2010 to about 10%in recent years,but the incidence rate is increasing annually.In 2017,the World Health Organization listed SFTS as one of the top ten key infectious diseases in urgent need of research.Hemorrhagic fever with renal syndrome(HFRS)is a natural focal disease caused by various viruses of hanta-viruses(HV)of Bunyaviridae family.It is mainly transmitted by rodents,with fever,congestive/hemorrhage,and renal damage as the main clinical manifestations.Severe patients may experience intractable shock and even multiple organ failure.With a globally epidemic of HFRS,China is a country with the worst epidemic situation.During the 1950s and 1970s,the mortality rate in China was as high as 14%,but it dropped to about 1%with the improvement of medical conditions.Etiological examination is the "gold standard" for the diagnosis of two diseases.The two diseases share similar epidemiological and clinical characteristics,and many primary hospitals lack the capacity to perform etiological examinations,which increases the risk of misdiagnosis or missed diagnosis in clinical settings.Objectives:This study aimed to compare the epidemiological and clinical characteristics of SFTS and HFRS patients and establish a differential diagnosis and prediction model based on clinical data.Methods:The subjects were laboratory-diagnosed inpatients with SFTS and HFRS at Shandong Public Health Clinical Center from January,2011,to June,2022.The study collected demographic data,clinical manifestations,laboratory and imaging findings,treatment,and disease prognosis of patients.The clinical characteristics of SFTS and HFRS patients were compared,and a risk prediction model for differential diagnosis was constructed.The Ethics Committee of Shandong Public Health Clinical Center approved this study.Statistical analysis was performed using SPSS 25.0 software.T-test or Mann-Whitney U test was employed for measurement data.The chi-square test or Fisher’s exact probability method was employed for classified data.P-values less than 0.05 were considered statistically significant.Indicators with statistical significance(P<0.05)in univariate analysis were included in multivariate logistic regression analysis.Results:1396 cases were enrolled,including 1166 cases of SFTS and 230 cases of HFRS.The median time from symptom onset until hospital admission in patients with SFTS and HFRS was the same,which was 7 days.The mortality of SFTS patients was 20.0%,and that of HFRS patients was 2.2%.The epidemiology characteristics,clinical manifestations,laboratory test results,and differential diagnosis prediction model of the two diseases are as follows.[Epidemiological characteristics]In this study,the number of SFTS inpatients increased year by year from 2011 to 2021,and the admission time was concentrated from spring to autumn(May to October).During 2011-2021,the number of HFRS inpatients showed a stable trend with little fluctuation,with the peak month of admission in autumn to winter(October-December).The number of patients with SFTS and HFRS admitted to the hospital in spring to autumn(April-November)was 1120(99.8%)vs 166(73.5%),respectively(P<0.001).The average ages of patients with SFTS and HFRS were 63.3± 11.2 years old and 46.3± 14.8 years old,respectively(P<0.001).The ratio of male to female patients with SFTS and HFRS was 1:1 and 5:1,respectively(P<0.001).928 patients(79.6%)with SFTS and 157 patients(68.3%)with HFRS lived in rural areas(P<0.001).[Clinical features]The common symptoms of SFTS patients were fever,anorexia,fatigue,lymphadenopathy,and nervous system symptoms,which were 1156 cases(99.1%),1078 cases(92.5%),979 cases(84.0%),839 cases(72.0%)and 545 cases(46.7%),respectively.The common symptoms of HFRS patients were fever,anorexia,headache/dizziness,fatigue,bleeding symptoms/signs,and vomiting,which were 229 cases(99.6%),205 cases(89.1%),158 cases(68.7%),138 cases(60.0%)and 105 cases(45.7%),respectively.In comparison,more patients with SFTS had lymphadenopathy and nervous system symptoms(P<0.001),and more patients with HFRS had bleeding symptoms/signs(P=0.001).[Laboratory examination]The median values of White blood cell count(WBC)(×109/L),Urea nitrogen(BUN)(mmol/L),Creatinine(Cr)(μmol/L),and Creatine kinase(CK)(U/L)in patients with SFTS were 2.9,5.1,68.0 and 393.0,respectively,while those in patients with HFRS were 11.0,13.6,190.9 and 101.0,respectively.The WBC,BUN and Cr values of HFRS patients were higher than those of SFTS patients(P<0.001),and CK values of SFTS patients were higher than those of HFRS patients(P<0.001).[Differential diagnosis risk prediction model of SFTS and HFRS]Univariate and multivariate analysis of clinical characteristics was performed on patients whose time from symptom onset until hospital admission was≤7 days and the equation of differential risk prediction model was formed as follows:P=ea/(1+ea),a=-5.86+3.22×age(≥60 years old)(0 or 1)+1.87×gender(female)(0 or 1)+4.95×(admission time from April to November)(0 or 1)+2.68 ×lymphadenopathy(0 or 1)+1.88 × neurological symptoms(0 or 1)+3.33×[WBC<(4 × 109/L)]/-3.00 ×[WBC>(10×109/L)(0,1 or 2)]-1.74×(ALT>40U/L)3.23×[Cr>98.1(female)/110(male)μmol/L](0 or 1)+3.25×(CK>174U/L)(0 or 1).The goodness-of-fit test of this model showed that P>0.999.The area under ROC curve(AUC)=0.996(95%CI:0.993-0.999,P<0.001),SD=0.002.The sensitivity was 97.0%and the specificity was 97.9%.According to the above risk prediction model,the risk prediction scoring criteria for the differential diagnosis of SFTS and HFRS were established:age(≥60 years old):3 points,gender(female):2 points,admission time from April to November:5 points,lymphadenopathy:3 points,nervous system symptoms:2 points,WBC<(4×109/L):3 points,WBC>(10×109/L):3 points,ALT>40U/L:-2 points,Cr>[98.1μmol/L(female)/110 μmol/L(male)]:-3 points,CK>174U/L:3 points.The lowest score was-8 points and the highest score was 21 points.According to the above risk prediction model formula and the scoring criteria,probability P of diagnosis as SFTS was calculated.There were three grades:low risk(-8-2 points)(P:0.000.05),medium risk(3-7 points)(P:0.06-0.80),and high risk(8-21 points)(P:0.81-1.00).The patients whose time from symptom onset until hospital admission was ≤7 days were graded according to this score,and the rate of diagnosis of SFTS in low risk,medium risk and high risk were 1.0%,32.3%,99.6%,respectively.Using SFTS cases with a time interval between symptom onset and hospital admission of greater than 7 days as the validation cohort,the predictive accuracy of the model was tested,and the AUC was 0.990(95%CI:0.982-0.997,P<0.001),SD=0.004.The sensitivity was 93.7%,and specificity was 97.2%.Using the scoring criteria,the diagnostic rates of SFTS in the lowrisk,medium-risk,and high-risk groups were 0.0%,52.7%,and 99.4%,respectively.Conclusions:The onset of SFTS primarily occurs from spring to autumn(May to October),with the majority of cases affecting the elderly population.The proportion between males and females were relatively similar.The typical clinical manifestations were lymphadenopathy and nervous system symptoms.The main laboratory findings were the decrease of WBC and the increase of CK.In contrast,HFRS had a peak incidence during autumn and winter(December to February)and predominantly affects middle-aged males.The typical clinical manifestations were bleeding symptoms/signs,and the main laboratory findings were the increase of WBC,BUN and Cr level.The risk prediction model of differential diagnosis between SFTS and HFRS was constructed based on the 9 indicators of age,gender,admission time,lymphadenopathy,nervous system symptoms,WBC,ALT,Cr,and CK,and the risk prediction scoring criteria for the differential diagnosis of SFTS and HFRS were determined:age(≥60 years old):3 points,gender(female):2 points,admission time from April to November:5 points,lymphadenopathy:3 points,nervous system symptoms:2 points,WBC<(4×109/L):3 points,WBC>(10 ×109/L):-3 points,ALT>40U/L:-2 points,Cr>[98.1μmol/L(female)/110 μmol/L(male)]:-3 points,CK>174U/L:3 points.According to the above risk prediction model formula,probability P of diagnosis as SFTS was calculated.There were three grades:low risk(-8-2 points)(P:0.000.05),medium risk(3-7 points)(P:0.06-0.80),and high risk(8-21 points)(P:0.81-1.00).The internal validation of prediction model and prediction scoring criteria demonstrated that had a good predictive effect.The study population was a single-center cohort from northern China,and further validation is needed to assess its generalizability.