Identification Of Prognostic Factors And Construction Of A Nomogram For Patients With Adult-onset Still’s Disease

Background and ObjectiveAdult-onset Still’s disease(AOSD)is a systemic inflammatory disease of unknown aetiology,characterised by leucocytosis,high fever,arthritis,rash,elevated liver enzymes,lymphadenopathy,hepatosplenomegaly,and multiple organ damage.Epidemiological investigation shows that the incidence of AOSD is about 2-3/1000,000.The incidence rate of female is higher than that of male.In addition,the age of onset is usually between 16 and 46 years old.With the continuous improvement of diagnostic technology.AOSD has been better identified,so the incidence has become an increasing trend in recent years.AOSD is considered to have multiple clinical outcomes.Cush et al.classified AOSD in 1987 as having three clinical outcomes,namely,polycyclic disease,chronic articular disease,and monocyclic disease.However,the study was based on a limited number of patients with the advancement in medical technology and the continuous development of new drugs,this classification was insufficient for the identification of the disease.In addition,there are only a few studies on prognostic factors of AOSD at present,mainly due to the unclear definition regarding clinical outcomes.Ruscitti et al.used system scores to define new disease outcomes to identify patients at a high risk of mortality;however,mortality is the least commonly noted event among the poor clinical outcomes associated with AOSD.Additionally,only a few studies have explored the poor factors associated with relapse,but they have ignored the patients with persistent,non-remission disease.The existence of this group can significantly affect the outcome.Thus far,there is also no clinical predictive model that can identify newly diagnosed patients as having good or poor prognoses.Therefore,the present study aimed to explore the clinical characteristics of AOSD and define comprehensive clinical outcomes,identify the clinical factors that affect the poor outcome of AOSD patients,and establish a well differentiated clinical nomogram to predict the individual risk of a poor prognosis in patients.Furthermore,it can promote a further understanding of the clinical characteristics of AOSD,which is conducive to the formulation of systematic and standard treatment strategies.Therefore,in view of this situation,in order to explore the clinical characteristics of AOSD and define comprehensive clinical outcomes,identify the clinical factors that affect the poor outcome of AOSD patients,and establish a well differentiated clinical nomogram to predict the individual risk of a poor prognosis in patients,We selected 174 patients who were treated in The First Affiliated Hospital of Zhengzhou University from January 2013 to February 2019 in our study.MethodsA total of 174 adult patients with AOSD who were newly diagnosed in the Department of Rheumatology and Immunology,the first affiliated Hospital of Zhengzhou University from January 2013 to February 2019 were included in this study.The general condition,clinical manifestation,laboratory examination,treatment and clinical outcome of AOSD were collected,and R software was used to analyze and visualize the data.Then the nomogram was established to predict the prognosis of the patients.Finally,receiver operating characteristic(ROC)curve analysis,calibration curves,and decision curve analysis(DCA)were used to assess the predictive ability of the nomograms.Results(1)During the follow-up period,90 patients(51.72%)were identified as having a good prognosis and 23 patients(13.22%)were identified as having refractory disease;of these 23,nine patients(5.17%)died due to disease-related causes.61 patients(35.06%)were identified as having a relapse,and 1 patient(0.57%)died from a disease-related cause.(2)28 clinical factors were included in the univariate logical regression.The results showed that seven clinical factors including age>=37,fever>=39.5℃,disease duration>=41 days,weight loss,platelets(PLT)>=253 ×109L,serum ferritin>=1700 ng/ml,and ESR<96mm/H were unfavourable factors for relapse/refractory AOSD(P<0.05).Subsequently,the above seven clinical factors were incorporated into multivariate binary logistic regression.The results finally showed age>=37,fever>=39.5℃ disease duration>=41 days,platelets(PLT)>=253×109L,serum ferritin>=1700ng/ml,and ESR<96mm/H as independent unfavourable factors for relapse/refractory AOSD(P<0.05).The analysis of two different subgroups of women and AOSD aged 16-46 years old showed that fever>=39.5℃ and disease duration>=41 days were both adverse prognostic factors.(3)Six variables including age,fever,disease duration,PLT,serum ferritin and ESR were selected to predict the occurrence of relapse/refractory AOSD.The area under ROC curve(AUC)of training set and validation set were 0.765 and 0.714 respectively.In addition,the calibration curve and DCA analysis showed a higher degree of model fit and clinical net income respectively,indicating that the model was constructed successfully.(4)Inadition,we evaluated the predictive effectiveness of our nomogram in women and AOSD aged 16-46 years old.ROC analysis,calibration curves,and DCA showed that our 6-factor nomogram had good predictive ability in both subgroups.(5)Although there was no statistically significant benefit from the use thereof,all four patients treated with tocilizumab achieved long-term remission.Conclusion:Novel nomograms based on clinical characteristics were developed and can be applied to help predict the individual risk of poor prognosis of patients.