Predictive value of peak systolic velocity for the development of graft limb complications after endovascular aneurysm repair.

Purpose: To determine the role of peak systolic velocity (PSV) data provided by duplex ultrasound (DUS) surveillance in the prediction of endograft limb complications after endovascular aneurysm repair (EVAR).

Methods: All 478 consecutive patients (425 men; mean age 75±7 years) who underwent infrarenal EVAR between 2004 and 2010 had DUS scans at 1.5, 3, 6, 9, 12, and 18 months and annually thereafter over a median follow-up of 43 months (range 1-92). In a retrospective study, the PSV recorded from the proximal and distal regions of each stent-graft limb was extracted from each postoperative DUS scan for each patient up to the penultimate scan before diagnosis of a limb complication (limb occlusion, symptomatic or hemodynamically significant kinking, or hemodynamically significant DUS-defined stenosis) requiring reintervention. The median (range) PSV readings from the proximal and distal regions of each stent-graft limb over the course of follow-up were compared between patients who developed a limb complication (n = 38) and those who did not (n = 440). Time-dependent Cox proportional hazards modeling was performed after risk adjustment; results are presented as the hazard ratio (HR) and 95% confidence interval (CI).

Results: In the proximal stent-graft limb segment, the median PSV was 106 cm/s (42-308) in patients without limb complications vs. 121 cm/s (50-281) in those with limb complications. Corresponding values in the distal segment of the endograft limb were 113 cm/s (35-400) vs. 129 cm/s (58-420). After risk adjustment, increased PSV over time within both the proximal and distal segments of the stent-graft limb was significantly associated with the risk of limb complications (proximal HR 1.015, 95% CI 1.003 to 1.028, p = 0.014; distal HR 1.010, 95% CI 1.001 to 1.020, p = 0.025).

Conclusion: Increases in the peak systolic velocity in stent-graft limbs were associated with an increased risk of limb complication, though no predictive threshold could be identified from scans prior to the development of a complication. This observation requires external validation and further investigation to define its clinical utility.