S10 The prognostic implications of acutely deteriorating interstitial lung disease in patients with idiopathic inflammatory myositis

• Published in: Thorax Vol. 78; no. Suppl 4; pp. A12 - A13
• Main Authors: Krishnan, V, Dobson, P, Aw, TC, Bandarage Chandramal, N, Molyneaux, PL, Kouranos, V, Donovan, J, Kokosi, M, George, P, Devaraj, A, Desai, S, Hewitt, R, Bax, S, Jenkins, GR, Alcada, J, Chua, F
• Format: Journal Article
• Language: English
• Published: London BMJ Publishing Group Ltd and British Thoracic Society 06.11.2023; BMJ Publishing Group LTD
• Subjects: Inflammatory diseases; Lung diseases; Mortality; Musculoskeletal diseases; Serology; ‘Survivor’ – Prognostic indicators in interstitial lung disease
• ISSN: 0040-6376; 1468-3296
• DOI: 10.1136/thorax-2023-BTSabstracts.16

IntroductionIdiopathic inflammatory myopathy-associated interstitial lung disease (IIM-ILD) can develop insidiously or manifest acutely, either as a fulminant initial presentation or unexpected deterioration of pre-existing ILD. In this study, we sought to characterise the disease course and prognostic determinants of acutely deteriorating IIM-ILD.MethodsThe Royal Brompton Hospital Immunology database was retrospectively interrogated for results of myositis-specific (MSA) and myositis-associated antibodies (MAA) from Jan 2015-Dec 2022. Patients with at least one acute IIM-ILD episode were identified using the SAS Clinical Data platform accessing Graphnet EPR and critical care systems. IIM were classified according to contemporaneous EULAR/ACR guidance and patients with <1 year follow-up were excluded. Time-to-death modelling was performed by Cox proportional hazards regression for acute and chronic (insidious) IIM-ILD and receiver operating curves generated to assess covariate prediction of mortality.Results344 patients (median age 61.2, IQR 50.8–69.6; 63.4% female and 63.1% White Caucasian) with acute and chronic IIM-ILD were followed up for a median of 4.5 years (IQR 2.4–8.7 years) and 5.5 years (IQR 3.7–8.4) respectively. MSA including Jo-1 antibody were present in 42.4% and 36.7% while MAA were detected in 78.0% and 80.5% of the acute and chronic subgroups respectively. ILD presentation preceded myositis in 57.8% of cases. Overall, a third (118/344; 34%) of patients were hospitalised for acute IIM-ILD, with a fifth requiring intensive care. 20.4% (68/344) of all patients died during follow-up. MDA5 serology was positive in five patients including two survivors of acute IIM-ILD. Unadjusted hazard ratios (HR) for death were highest for acute ILD deterioration, age >60 and elevated pulmonary arterial pressure. Multivariate regression confirmed that they independently predicted mortality; the highest adjusted HR (6.64, 95% CI:3.7–11.9; P<0.001) was attributed to acute IIM-ILD. The discriminatory index (AUC) for the integrated model was 0.84. Covariate stratification revealed tiered subgrouping with significant survival differences (figure 1).Abstract S10 Figure 1Estimated survival by covariate stratificationConclusionsIn identifying clinical variables that predict a poorer outcome in IIM, acute ILD development but not anti-Jo-1 serology was strongly associated with an increased likelihood of death. Detailed profiling including predictive stratification may allow high-risk patients with IIM-ILD to be triaged for earlier and more intensive intervention.