SAT0526 COMPARISON OF INFLAMMATORY MARKERS AS DIAGNOSTIC TOOL IN PATIENTS WITH ADULT-ONSET STILL’S DISEASE

Abstract

Background:The diagnosis for adult-onset Still’s disease (AOSD) is still based on nonspecific symptoms and laboratory data, and had necessity to rule out several infectious, autoimmune or malignant diseases.Objectives:This study aimed to elucidate the efficiency of inflammatory markers, including systemic immune-inflammation index (SII); C-reactive protein (CRP)-to-albumin ratio (CAR), albumin-to- glubulin ratio (AGR), prognostic nutritional index (PNI), and ferritin-to-ESR ratio (FER) for evaluation of diagnostic or prognostic factors in AOSD.Methods:The medical records of patients with suspected AOSD between January 1999 and June 2019 were collected and retrospectively analyzed. Among 225 patients, 61 patients received another diagnosis, such as infection and malignancy, and 164 patients were newly diagnosed to AOSD. The values of SII, CAR, AGR, PNI, and FER were compared with AOSD and non-AOSD groups. Receiver operating characteristic (ROC) curve analysis was performed to evaluate the diagnostic significance of inflammatory markers. Correlations between inflammatory markers and disease activity index were analyzed.Results:A total of 164 patients who diagnosed AOSD had higher values of SII, CAR, and FER as well as lower values of AGR and PNI. For AOSD diagnosis, the area under the curve (AUC) obtained from the ROC curve were 0.859 (95% CI=0.806-0.911) for SII, 0.769 (95% CI=0.702-0.837) for CAR, 0.749 (95% CI=0.615-0.782) for AGR, 0.699 (95% CI=0.675-0.823) for PNI, and 0.764 (95% CI=0.693-0.834) for FER with cut-off value of 2195.7, 1.80, 1.38, 48.8 and 17.0, respectively. The SII had the largest AUC, and FER and SII each had the highest sensitivity (70.9%) and specificity (91.5%). In correlation analysis, there were no strong correlations between inflammatory markers and disease activity indices except CAR and CRP.Table 1.Demographic and baseline characteristics of patients with adult-onset Still’s disease(AOSD)AOSD patients (N=164)Initially suspected AOSD but not diagnosed (N=61)P-valueAge, years42.6 ± 15.742.1 ± 14.40.805Fever162 (98.8)56 (91.8)0.017Sore throat93 (56.7)18 (29.5)<0.001Skin rash131 (79.9)26 (42.6)<0.001Lymphadenopathy63 (38.4)10 (16.4)0.002Hepatomegaly25 (15.2)1 (1.6)0.004Arthritis79 (48.2)22 (36.1)0.132SII4601.7 ± 6406.81105.6 ± 970.0<0.001CAR3.37 ± 2.821.22 ± 1.58<0.001AGR1.15 ± 0.351.38 ± 0.39<0.001PNI41.94 ± 6.7048.85 ± 8.43<0.001FER139.19 ± 372.1228.39 ± 70.800.006Table 2.Discriminatory ability of inflammatory markers for predicting AOSD diagnosisVariablesAUCP-valueCut-offSensitivitySpecificity+LR-LRSII0.859<0.0012195.764.6%91.5%7.600.39CAR0.769<0.0011.8065.2%74.6%2.570.47AGR0.700<0.0011.3854.1%82.3%3.060.56PNI0.752<0.00148.852.5%86.0%3.750.55FER0.7640.00617.070.9%74.6%2.760.39Figure 1.Receiver operating characteristic (ROC) curves for SII, CAR and FER in AOSD and non-AOSD patients (Initially suspected AOSD but not diagnosed).Figure 2.Receiver operating characteristic (ROC) curves for AGR and PNI in AOSD and non-AOSD patients (Initially suspected AOSD but not diagnosed).Conclusion:SII, CAR, AGR, PNI and FER were can be used as a practical tool for diagnosing AOSD. Among the inflammatory markers, SII was revealed as most powerful marker for diagnosis.