Cardiac magnetic resonance predicts ventricular arrhythmias in scleroderma: the Scleroderma Arrhythmia Clinical Utility Study (SAnCtUS). (25th November 2019)

• Record Type: Journal Article
• Authors: Mavrogeni, Sophie; Gargani, Luna; Pepe, Alessia; Monti, Lorenzo; Markousis-Mavrogenis, George; De Santis, Maria; De Marchi, Daniele; Koutsogeorgopoulou, Loukia; Karabela, Georgia; Stavropoulos, Efthymios; Katsifis, Gikas; Bratis, Konstantinos; Bellando-Randone, Silvia; Guiducci, Serena; Bruni, Cosimo; Moggi-Pignone, Alberto; Dimitroulas, Theodoros; Kolovou, Genovefa; Bournia, Vasiliki-Kalliopi; Sfikakis, Petros P; Matucci-Cerinic, Marco
• Is Part Of: Rheumatology. Volume 59:Number 8(2020)
• Page Start: 1938
• Page End: 1948
• Publication Date: 2019-11-25
• Subjects: scleroderma -- systemic sclerosis -- cardiovascular magnetic resonance -- sudden cardiac death -- rhythm disturbance; Rheumatism -- Periodicals; Rheumatology -- Periodicals; 616.723005
• DOI: 10.1093/rheumatology/kez494
• Languages: English
• ISSNs: 1462-0324

Abstract

Abstract: Objectives: Cardiac rhythm disturbances constitute the most frequent cardiovascular cause of death in SSc. However, electrocardiographic findings are not a part of risk stratification in SSc. We aimed to translate 24 h Holter findings into a tangible risk prediction score using cardiovascular magnetic resonance. Methods: The Scleroderma Arrhythmia Clinical Utility Study (SAnCtUS) was a prospective multicentre study including 150 consecutive SSc patients from eight European centres, assessed with 24 h Holter and cardiovascular magnetic resonance, including ventricular function, oedema (T2 ratio) and late gadolinium enhancement (%LGE). Laboratory/clinical parameters were included in multivariable corrections. A combined endpoint of sustained ventricular tachycardia requiring hospitalization and sudden cardiac death at a median (interquartile range) follow-up of 1 (1.0–1.4) year was generated. Results: Only T2 ratio and %LGE were significant predictors of ventricular rhythm disturbances, but not of supraventricular rhythm disturbances, after multivariable correction and adjustment for multiple comparisons. Using decision-tree analysis, we created the SAnCtUS score, a four-category scoring system based on T2 ratio and %LGE, for identifying SSc patients at high risk of experiencing ventricular rhythm disturbance at baseline. Increasing SAnCtUS scores were associated with a greater disease and arrhythmic burden. All cases of non-sustained ventricular tachycardia ( n = 7) occurred in patients with the highest SAnCtUS score (=4). Having a score of 4 conveyed a higher risk of reaching the combined endpoint in multivariable Cox regression compared with scores 1/2/3 [hazard ratio (95% CI): 3.86 (1.14, 13.04), P = 0.029] independently of left ventricular ejection fraction and baseline ventricular tachycardia occurrence. Conclusion: T2 ratio and %LGE had the greatest utility as independent predictors of rhythm disturbances in SSc patients.