POS0912 SARCOPENIA IN SPONDYLOARTHRITIS. ULTRASOUND EVALUATION

Abstract

BackgroundSpondyloarthritis are a group of chronic inflammatory rheumatic diseases that share clinical and imaging characteristics. Sarcopenia is characterized by loss of muscle mass and function associated with impairment of quality of life and adverse outcomes such as fractures and death. Patients with Spondyloarthritis had a prevalence of up to 25.7% of sarcopenia. The available imaging techniques for the detection of low muscle quantity and quality are: dual-energy X-ray absorptiometry (DXA), bioelectrical impedance analysis (BIA), and recently muscular ultrasound (US). Ultrasound estimate muscle quantity using prediction equations based on muscle thickness that have shown to be highly reliable in healthy population. Among the models used, the following equations are applicable in different ethnicities to predict muscle mass Abe 2018, Barbosa Silva 2021, Tang 2022.ObjectivesTo evaluate the prevalence of sarcopenia in Spondyloarthritis and to determine which of the three ultrasonographic predictive models correlates best with BIA and DXA in our population.MethodsObservational, cross-sectional study. We included 42 patients with spondyloarthritis (according to ASAS criteria for ankylosing spondylitis or CASPAR criteria for psoriatic arthritis). Appendicular skeletal muscle mass (ASMM) was evaluated by DXA and predicted by BIA and US. We performed a US examination with a GE LOGIQTMe, equipped with a broadband linear probe to measure muscle thickness (MT) at seven sites. Based on ultrasound prediction models, we use age, sex, height, and MT to estimate ASMM (Abe-2018, Barbosa-2021, and Tang-2022). A bivariate analysis was performed to assess the differences between patients with and without decreased ASMM. Pearson´s correlation coefficient was calculated to evaluate the correlation of the ultrasonographic predictive models with DXA and BIA. Finally, a ROC curve analysis was performed to determine the area under the curve of the models.ResultsWe included 42 patients, composed mainly of men (24/42, 57.1%), with ankylosing spondylitis (AS) in 35 (83.3%) and psoriatic arthritis in 7 (16.7%). The prevalence of sarcopenia was about 33% (10/14 sarcopenia y 4/14 severe sarcopenia). When comparing patients with and without decreased ASMM, differences were only found in ultrasonography measurements of the quadriceps and anterior tibial muscles (p=0.003 y 0.00). On the other hand, the ultrasound prediction models had a moderate to strong correlation with DXA and BIA (0.80, 0.87, 0.53 and 0.88, 0.74, 0.66 for Abe-2018, Barbosa-2021, and Tang-2022). Nevertheless, in the ROC curve, the ABE-2018 model expresses a better diagnostic performance (AUC 0.80 p=0.002).ConclusionResults indicated a low ASMM in 30% of the patients with Spondyloarthritis (sarcopenia 24.4%, severe sarcopenia 9.8%). This study supports the use of ultrasound and shows a good correlation with DXA and BIA for ASMM evaluation in Spondyloarthritis, particularly with the Abe-2018 predictive model.Reference[1]Van den Broeck J, Buzzatti L, Jager-Wittenaar H, Perkisas S, Scafoglieri A. The validity of ultrasound-derived equation models to predict whole-body muscle mass: A systematic review. Clin Nutr ESPEN. 2021; 46:133–41.https://clinicalnutritionespen.com/article/S2405-4577(21)00300-4/fulltextTable 1.General characteristics of patientsVariableNormal ASMMN (%)Low ASMMN (%)*pMen14 (58.3)10 (41.6).22AS21 (60)14 (40).19Age48.85+12.6749.92+17.38.82HLA-B2716 (66.7)9 (90).13Arthritis8(66.6)4(33.3).93Enthesis6 (75)2 (25).60Low grip strength8(30.8)4 (33.3).87Gait speed 6 m <1 m/s16 (61.5)10 (83.3).16Ultrasound†‡QuadricepsTibialis anterior3.81+.524.67+.453.47+.463.97+.46.00.00Predictive models US†*ABE 2018BARBOSA 2021TANG 20217.36+1.027.91+.958.72+.606.58+.677.56+.927.72+.63.015.268.000AS, Ankylosing spondylitis; PsA, Psoriatic arthritis;†= Mean, standard deviation;‡= Presented in cm; *= Presented in kg/m2, IQR= Interquartile rangeGraph 1.Area under de curveAcknowledgements:NIL.Disclosure of InterestsNone Declared.