PO:06:090 | Validation of 'Clinical Red Flags' for RA-ILD in an Italian cohort: the role fo Dyspnea, Cough and Crackles

Eneida Cela¹, Chiara Bonini¹, Francesco Cavalli², Alessandra Luciano³, Barbara Kroegler¹, Josuel Ora², Alberto Bergamini¹, Marcello Chiocchi³, Paola Rogliani², Paola Conigliaro¹, Maria Sole Chimenti¹ | ¹Rheumatology, Allergology and Clinical Immunology, Department of System Medicine, University of Rome Tor Vergata, Rome, Italy; ²Division of Respiratory Medicine, University Hospital Policlinico Tor Vergata, Rome, Italy; ³Department of Diagnostic Imaging and Interventional Radiology, University of Rome, Tor Vergata, Rome, Italy

Background. Rheumatoid arthritis (RA) is a systemic inflammatory disease, with articular and extra-articular manifestations, including Interstitial Lung Disease (ILD). ILD has important clinical implications, making early diagnosis essential. Previously, an Italian study, through a Nominal Group Technique and a Delphi Survey identified a “checklist for clinical practice”, including red flags of signs and symptoms suggestive of RA-ILD1. We aimed to validate the “checklists of red flags signs and symptoms” in our cohort of RA patients, by assessing the diagnostic utility of dyspnea, cough, and crackles, both individually and in combination.

Methods. We performed a retrospective study of RA patients charts from a Rheumatologic outpatient clinic (May 2022 -June 2024). Inclusion criteria: 1) RA diagnosis from at least 6 months; 2) Thoracic High-Resolution CT (HRCT) during follow-up. Exclusion: 1) Incomplete follow-up data; 2) Missing pulmonary exam details. RA-ILD diagnosis was based on a blind evaluation of the HRCTs by two radiology specialists and through a clinical assessment by two pneumology specialists. Symptoms and signs of ILD (cough, crackles, and dyspnea) were considered separately and combined to determine sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV). Statistical significance was assessed via Chi-square or Fisher's exact test (p < 0.05).

Results. We analysed 107 medical records of RA patients, 55 (51.4%) were diagnosed with RA-ILD. Female patients were prevalent in both ILD and No-ILD groups (56.4% and 82.6% respectively) (Table 1). The mean age at ILD diagnosis was 67.4 ± 10.4 years. Table 2 shows the sensitivity, specificity, PPV, NPV, accuracy and p-value of the red flags. Dyspnea showed good diagnostic performance: sensitivity 63.5%, specificity 60%, PPV 60%, NPV 63.5% (p = 0.0203). Crackles had the highest sensitivity (66.7%) and NPV (75%), with specificity of 57.4% (p = 0.0265). Cough showed the weakest diagnostic performance. Combined analysis revealed that dyspnea or crackles had the strongest association with RA-ILD (accuracy 63.5%, specificity 63.8%, p = 0.0066). Cough or dyspnea also showed significant association (p = 0.0111), while cough or crackles (p = 0.2481) and cough with dyspnea (p = 0.3202) did not. A combination of both crackles and dyspnea correlated significantly with RA-ILD (p = 0.0351). Cough with crackles showed a trend toward significance (p = 0.0927). When all three symptoms were present, sensitivity was 64.3%, specificity 53.2%, PPV 32.7%, NPV 81%, but the result lacked significance (p = 0.1284).

Conclusions. Crackles and dyspnea are reliable markers for RA-ILD, both alone, and in combination. This study confirms the diagnostic value of the red flags previously identified, especially in combination, in the early detection of RA-ILD. 1. Bosello, S. L. et al. Interstitial Lung Disease Associated With Autoimmune Rheumatic Diseases: Checklists for Clinical Practice. Front. Med. 8, (2021).