Psoas Muscle Area as a Prognostic Factor for Survival in Patients Undergoing Endovascular Aneurysm Repair Conversion

Endovascular aneurysm repair conversion (EVAR-c) is increasingly reported and known to be technically complex and physiologically demanding. It has been proposed that pragmatic anthropomorphic measures such as psoas muscle area (PMA) may reliably quantify levels of preoperative frailty and be used to inform point of care clinical decision-making and patient discussions for a variety of complex operations. To date, there is mixed data supporting use of PMA as a prognostic factor in fenestrated endovascular and open abdominal aortic aneurysms (AAA) repairs; however, no literature exists evaluating the impact of preoperative PMA on EVAR-c results. Therefore, the purpose of this study was to review our EVAR-c experience and evaluate the association of PMA with perioperative and long-term mortality outcomes.

A retrospective single-center review of all AAA repairs was performed (2002-2019) and EVAR-c procedures were subsequently analyzed (n = 153). Cross-sectional PMA at the mid-body of the L3 vertebrae was measured. The lowest PMA tertile was used as a threshold value to designate patients as having "low" PMA (n = 51) and this cohort was subsequently compared to subjects with "normal" PMA (n = 102). Cox proportional hazards modeling was used to estimate covariate association with all-cause mortality.

Patients with low PMA were older (77 vs. 72 years; P = 0.002), more likely to be female (27% vs. 5%; P < 0.001), and had reduced body mass index (26 vs. 29 kg/m²; P = 0.002). Time to conversion, total number of endovascular aneurysm repair (EVAR) reinterventions prior to conversion and elective EVAR-c presentation incidence were similar; however, patients with low PMA had larger aneurysms (8.3 vs. 7.5 cm; P = 0.01) and increased post-EVAR sac growth (2.3 vs. 1 cm; P = 0.005). Unadjusted inpatient mortality was significantly greater for low PMA patients (16% vs. normal PMA, 5%, P = 0.02). Similarly, the total number of complications was higher among low PMA subjects (1.5 ± 1.9 vs. normal PMA, 0.9 ± 1.5; P = 0.02). Although frequency of major adverse cardiovascular events and new onset inpatient hemodialysis were similar, low PMA patients had a more than four-fold increased likelihood of having persistent requirement of hemodialysis at discharge (18% vs. 4%,P = 0.01). The low PMA group had decreased survival at 1 and 5 years, respectively (77 ± 5%, 65 ± 6% vs. normal PMA, 86 ± 3%, 82% ± 5%; log-rank P = 0.03). Low PMA was an independent predictor of mortality with every 100 mm² increase in PMA being associated with a 15% reduction in mortality (hazard ratio: 0.85,95% confidence interval:, 0.74-0.97; P = 0.02).

Among EVAR-c patients, subjects with low preoperative PMA had higher rates of postoperative complications and worse overall survival. PMA assessments may be a useful adjunct to supplement traditional risk-stratification strategies when patients are being considered for EVAR-c.