Computed tomography angiography-based evaluation of great saphenous vein conduit for lower extremity bypass

Objective

Lower extremity computed tomography angiography (CTA) is frequently used for anatomic assessment of lower extremity peripheral arterial disease. When lower extremity bypass is planned, duplex ultrasound (DUS) is routinely obtained to evaluate the great saphenous vein (GSV) for use as conduit. Although GSV can be visualized on CTA images, diameter assessment is not routinely included in formal study interpretation. We hypothesized that CTA images could be used to measure GSV diameters and that CTA-based diameters would correlate with measurements obtained using DUS.

Methods

Consecutive patients undergoing lower extremity arterial bypass who were evaluated preoperatively with both CTA and DUS vein mapping were identified at a single hospital. Minimum above- and below-knee GSV diameters were measured from electronically archived CTA images by two independent observers. CTAs were performed using standard arterial phase protocol without additional venous phase imaging. Between-observer reproducibility of CTA-based diameter measurements was evaluated using intraclass correlation coefficients. Correlation between CTA and DUS-based GSV diameters was evaluated with Spearman correlation coefficients. CTA diameter cut-points for identification of adequate GSV bypass conduit, defined as DUS-based minimum GSV diameter ≥ 3 mm, were determined using receiver-operating characteristic curves.

Results

Sixty-three lower extremities were evaluated in 36 patients. In the absence of previous surgical removal, GSV was visible on all CTAs reviewed. No instances of GSV thrombosis were identified on DUS. Minimum DUS-based above-knee GSV diameter was 2.9 ± 0.1 mm (range, 1.4-4.6 mm), and mean below-knee diameter was 2.6 ± 0.1 mm (range, 1.3-4.0 mm). When GSV was visible and exceeded the minimum diameter threshold for CTA measurement, correlation between CTA- and DUS-based diameters was both positive and highly significant (ρ = 0.595; P < .0001). CTA-based diameters also had excellent reliability between observers (r [95% CI]: 0.88 [0.85-0.91]). For identification of adequate bypass conduit using CTA, above-knee GSV diameter ≥ 3.9 mm was 67% sensitive and 73% specific; below-knee GSV diameter ≥3.0 mm was 75% sensitive and 84% specific.

Conclusions

CTA-based GSV diameter measurements have good reproducibility and highly significant correlation with DUS-based diameters. CTA-based GSV diameter is a specific but relatively insensitive indicator of adequate bypass conduit. When CTA-based diameters indicate inadequate GSV bypass conduit, confirmatory DUS vein mapping is warranted. Confirmatory DUS vein mapping may be unnecessary when adequate vein diameter is identified on CTA.