Left Ventricular Systolic Myocardial Function in Ankylosing Spondylitis

• Published in: Arthritis care & research (2010) 2019-09, Vol.71 (9), p.1276-1283
• Main Authors: Midtbø, Helga, Semb, Anne G., Matre, Knut, Rollefstad, Silvia, Berg, Inger J., Gerdts, Eva
• Format: Article
• Language: English
• Subjects: Adult; Age Factors; Ankylosing spondylitis; Aorta; Body mass index; Cardiovascular diseases; Cardiovascular Diseases - diagnostic imaging; Cardiovascular Diseases - etiology; Cardiovascular Diseases - physiopathology; Case-Control Studies; Echocardiography; Echocardiography, Doppler - methods; Female; Health risk assessment; Heart; Humans; Image Interpretation, Computer-Assisted; Incidence; Linear Models; Male; Middle Aged; Multivariate Analysis; Myocardial Contraction - physiology; Observer Variation; Population studies; Prognosis; Retrospective Studies; Risk Assessment; Risk factors; Severity of Illness Index; Sex Factors; Spondylitis; Spondylitis, Ankylosing - complications; Spondylitis, Ankylosing - diagnosis; Ventricle; Ventricular Dysfunction, Left - diagnostic imaging; Ventricular Dysfunction, Left - epidemiology; Ventricular Dysfunction, Left - etiology; Ventricular Function, Left - physiology
• ISSN: 2151-464X; 2151-4658
• DOI: 10.1002/acr.23765

Objective Subclinical left ventricular (LV) myocardial dysfunction is associated with an increased risk of cardiovascular disease (CVD), but it is not known whether subclinical LV myocardial dysfunction is present in patients with ankylosing spondylitis (AS) independent of CVD risk factors. Methods Conventional and speckle tracking echocardiography were performed in 106 patients with AS (mean ± SD age 48 ± 12 years; 59% men) and 106 matched controls (mean ± SD age 51 ± 12 years; 59% men). LV systolic myocardial function was assessed by peak systolic global longitudinal strain (GLS). Results CVD risk factors were similarly distributed in patients with AS and controls, but more controls received statin therapy (P = 0.05). GLS was significantly lower in patients with AS compared to controls (mean ± SD −17.7 ± 2.5% versus −18.4 ± 2.3%; P = 0.03). In univariable linear regression analyses in the total study population, lower GLS was associated with having AS, male sex, higher body mass index, higher LV mass index, and lower LV ejection fraction (all P < 0.05). Having AS retained an independent association with lower GLS when adjusted for these factors in multivariable analyses (β = 0.16, P = 0.02). In patients with AS, lower GLS was independently associated with larger aortic root diameter in multivariable analyses (β = 0.24, P = 0.02), while no association with AS disease activity, disease duration, or use of antirheumatic medication was observed. Conclusion Patients with AS had lower GLS compared with controls, independent of confounders. In AS patients, lower GLS was associated with larger aortic root diameter. Prospective studies should test whether lower GLS contributes to the observed higher CVD risk in patients with AS.