Cardiac Symptoms Are A Survival Predictor In Sarcoidosis

Aim: To correlate presenting symptoms with a final diagnosis of cardiac sarcoid & determine prognostic characteristics for mortality Methods: Retrospective analysis of 503 consecutive patients with sarcoidosis presenting between 2006-2016 to Papworth & Cambridge University Hospitals. Cardiac symptoms (specifically syncope, palpitations, chest pain) were correlated with diagnosis of cardiac sarcoidosis. Transplant-free survival analysis was performed. Prognostic markers were identified using uni- & multivariable logistic regression. Results: 27 patients (5.4%) were diagnosed with cardiac involvement. Presentation with cardiac symptoms was correlated with a diagnosis of cardiac sarcoid (p=0.006) though the overall predictive value was poor with a ROC of 0.62 (95% CI 0.51-0.73) due to a high false negative rate (65%) & a false positive rate of 10%. A diagnosis of cardiac involvement carried no significant increase in mortality/transplant rate compared with non-cardiac sarcoid (7.4% vs 6.3%). It was not associated with a worse transplant-free survival probability in multivariate regression (adjusted for age, stage, and TLCO % predicted) whereas a presentation with cardiac symptoms adjusted for the same confounders was (HR 4.8, 95% CI 1.2-19.2, p=0.03) Inclusion of pulmonary hypertension in the model did not change overall outcome. Conclusion: Cardiac symptoms are an important but unreliable marker for cardiac sarcoidosis. This series9 poor specificity highlights the importance of cardiac screening. The fact that mortality is comparable to non-cardiac cases is a testament to treatment advances. Cardiac symptoms at presentation do however carry a poor prognosis possibly due to being a marker of other underlying cardiac morbidity.