AN ULTRASOUND NEGATIVE FOR SUBCLINICAL SYNOVITIS IN PATIENTS WITH ARTHRALGIA: IS IT HELPFUL IN IDENTIFYING THOSE WHO WILL NOT DEVELOP INFLAMMATORY ARTHRITIS? A LONGITUDINAL STUDY IN FOUR ARTHRALGIA COHORTS

Background: Ultrasound (US) has become an established method in the evaluation of joints and is often used in clinical practice to guide management decisions in arthralgia-patients. To date, studies on the prognostic value of MSUS in arthralgia focused on the positive predictive value of subclinical inflammation. However, also the absence of imaging-detected subclinical synovitis is now increasingly used in daily practice to exclude arthralgia-patients from further follow-up. Though, evidence on the value of a negative US in ruling out future IA development in arthralgia-patients (the negative predictive value) is mostly absent. According to the rules of Bayes, predictive values are highly dependent on the prior-risk of developing the disease. The NPV therefore, is strongly related to the prior-risk of not getting IA, which is quite considerable in arthralgia-patients. Objectives: To investigate the negative predictive value (NPV) of musculoskeletal ultrasound (MSUS) in arthralgia patients at risk for developing inflammatory arthritis (IA). Methods: An MSUS examination of hands and feet was performed in arthralgia-patients at risk for IA in four independent cohorts. Patients were followed for one-year on the development of IA. Subclinical synovitis was defined as greyscale≥2 and/or power Doppler≥1. NPVs were determined and compared with the prior risks of not developing IA. Outcomes were pooled using meta-analyses and meta-regression analyses. In sensitivity analyses, MSUS-imaging of tender joints only (rather than the full US-protocol) was analyzed and ACPA-stratification applied, the latter being in line with the use of US in daily care. Results: After one-year 78%, 82%, 77% and 72% of patients in the four cohorts did not develop IA. The NPV of a negative US was 86%, 85%, 82% and 90%, respectively. The meta-analysis showed a pooled non-IA prevalence of 79%(95%CI: 75%-83%) and a pooled NPV of 86%(95%CI:81-89%) ( Figure 1 ). Imaging tender joints only (as generally done in clinical practice) and ACPA-stratification showed similar results. Figure 1. Full US protocol; Prior risks of not developing IA (A) and negative predictive values of MSUS (B) in the four cohorts. For comparison, the pooled prior risk and confidence interval from A are depicted in the red column in B. Conclusion: A negative US result in arthralgia has a high NPV for not developing IA, which is mainly due to the high a-priori risk of not developing IA. The added value of a negative US (<10% increase) was limited. REFERENCES: N.A. Disclosure of Interests: Cleo Rogier: None declared, Giulia Frazzei: None declared, Marion Kortekaas: None declared, Marloes Verstappen: None declared, Sarah Ohrndorf: None declared, Elise van Mulligen: None declared, Ronald van Vollenhoven Speakers bureau: Speaker, for which institutional and/or personal honoraria were received: AbbVie, Galapagos, GSK, Janssen, Pfizer, UCB, Consultant of: Consultancy, for which institutional and/or personal honoraria were received: AbbVie, AstraZeneca, Biogen, Biotest, BMS, Galapagos, Gilead, Janssen, Pfizer, Sanofi, Servier, UCB, Vielabio, Grant/research support from: Research Support (institutional grants): BMS, GSK, Lilly, UCB Support for Educational programs (institutional grants): Pfizer, Roche , Dirkjan van Schaardenburg: None declared, Pascal de Jong: None declared, Annette van der Helm-van Mil: None declared Citation: , volume 81, supplement 1, year 2022, page 1045Session: Epidemiology, risk factors for disease or disease progression (POSTERS only)