Duplex criteria for predicting progression of vein graft lesions: Which stenoses can be followed?

Serial color duplex imaging (CDI) is a sensitive technique for detecting infrainguinal vein graft stenosis. Flow disturbance sites can be accurately localized and degree of stenosis graded using peak systolic velocity (Vp in cm/sec) and velocity ratio (Vr) measurements. While minimal criteria for graft stenosis include Vp > 150 and Vr > 1.5, agreement upon threshold criteria useful in predicting progression to a graft-threatening stenosis is lacking. From August 1991 to April 1994, 61 infrainguinal vein bypass grafts in 57 male patients (mean age 63.4 years) underwent prospective duplex surveillance. CDI (5-10 MHz linear array, spectral analysis 4-6 MHz) was performed at 1 and 6 weeks, and then 3, 6, and every 6 months thereafter. Mean follow-up was 12.9 months (range 2.0-30.0). Thirty-nine stenoses were detected in 27 grafts. Fourteen stenoses in 12 grafts resolved spontaneously (mean Vp = 202, range 158-250; mean Vr = 2.3, range 1.4-3.7). Ten of these stenoses regressed within 3 months, while 4 resolved over 5-11 months. Nine lesions persisted during follow-up (mean Vp = 272; mean Vr = 3.9). In contrast, 16 stenoses (11 grafts) required revision (mean VP = 339.5, range 170-578; mean Vr = 7.0, range 1.7-19.9). Progression to a significant stenosis occurred within 4 months of its detection in 75% (12/16). Revised lesions had a higher Vp and Vr than those that spontaneously resolved (p < 0.05). By scatterplot analysis, a Vp > 300 and Vr > 3.5 differentiated lesions prone to progress from those likely to stabilize or resolve. Serial CDI measurements of Vp and Vr allow detection and stratification of vein graft stenoses. Surveillance of identified stenoses may be required every 6 weeks to detect evidence of rapid progression. Lesions reaching threshold values of Vp > 300 and Vr = 3.5 are unlikely to regress and should be considered for prophylactic revision.