RESUMO
BACKGROUND: Multiple CT-derived measurements of sarcopenia have been described yet their relationship with survival after abdominal aortic aneurysm (AAA) repair has not been properly assessed. We aimed to define and compare the relationship between several psoas CT-derived measurements and the 5-year survival after AAA repair and to evaluate their potential contribution to survival prediction. METHODS: Preoperative CT area (TPA) and density (MTPD) of the psoas muscle at L3 were measured in 218 consecutive AAA patients electively intervened. Additional measurements were obtained by normalizing TPA by anthropometric data or L3-vertebra surface or by TPAxMTPD multiplication (lean psoas muscle area-LPMA). The association of sarcopenia markers with survival was evaluated with Cox models adjusted by age, sex, type of intervention and the Charlson Comorbidity Index, and their contribution to survival prediction assessed with the C-statistic and the Continuous Net Reclassification Index (c-NRI). RESULTS: Sixty patients (27.5%) died during the first 5 years after surgery. There was a statistically significant and linear (spline analysis) relationship of sarcopenia markers with 5-year survival in all multivariate models, except that including LPMA. Despite this association, the inclusion of sarcopenia markers did not improve the C-statistic and moderately increased the c-NRI. None normalized sarcopenia markers performed better than TPA. CONCLUSION: The majority of CT-derived psoas muscle measurements of sarcopenia showed a significant and independent relationship with survival after elective AAA repair. Despite this association, they did not appear to improve sufficiently our survival prediction ability to become an efficient tool for decision-making.