Accurate and convenient methods for assessing a patient's risk of postoperative morbidity and mortality comprise important tools in clinical decision-making. Whereas some aspects of the patient's fitness for surgery can be easily quantified, measurement of the patient's frailty is often difficult or time-consuming. Previous research in the context of multiple types of major surgical procedures has reported psoas-L4 vertebral index (PLVI) to be a useful predictor of postoperative morbidity and mortality.
This retrospective cohort study assessed the hypothesis that PLVI can predict amputation-free survival (AFS) in patients undergoing open or endovascular lower extremity revascularization. The records of all lower extremity revascularization patients with preoperative computed tomography arteriography before revascularization during a recent 6-year period were reviewed for demographic information and outcomes. With use of embedded computed tomography software, the cross-sectional area of the bilateral psoas muscles and vertebral body at the L4 level were measured and used to calculate the PLVI. Univariate, multivariate logistic regression, and Cox proportional hazards analyses were performed for the primary outcome of AFS.
During a 6-year period, 188 patents had preoperative scanning, qualifying for inclusion in the study; 52% received open surgical bypass and 48% received a percutaneous endovascular procedure, with a median duration of follow-up of 12 months (interquartile range [IQR], 3-24 months). Median bilateral psoas cross-sectional area was 24.9 cm2 (IQR, 20.5-29.7 cm2), and mean PLVI was 1.74 (IQR, 1.39-2.05). Cox proportional hazards analysis identified age (hazard ratio [HR], 1.07; 95% confidence interval [CI], 1.01-1.14; P = .026), congestive heart failure (HR, 4.7; 95% CI, 1.29-16.9; P = .019), and dyslipidemia (HR, 0.34; 95% CI, 0.12-0.99; P = .049) as independent predictors of AFS loss, whereas PLVI was not (HR, 2.6; 95% CI, 0.83-8.39; P = .099). Kaplan-Meier life-table analysis demonstrated no significant differences in survival between the highest and lowest PLVI cohorts of patients. Hazard analysis showed concomitant congestive heart failure (HR, 15; 95% CI, 1.1-210; P = .042) and serum albumin concentration (HR, 0.16; 95% CI, 0.05-0.52; P = .0026) to be independent predictors of limb loss, whereas advanced age (HR, 1.20; 95% CI, 1.07-1.35; P = .0026), bypass procedure (HR, 4.6; 95% CI, 1.04-21; P = .045), non-African American race (HR, 9.09; 95% CI, 1.02-100; P = .048), and higher PLVI (HR, 10.9; 95% CI, 1.7-72; P = .013) predicted increased risk of mortality.
PLVI did not predict AFS after intervention for peripheral arterial occlusive disease. This is contrary to the ability of PLVI to predict perioperative and midterm survival after abdominal aortic aneurysm repair and other major abdominal surgery.