The correlation of early flow disturbances with the development of infrainguinal graft stenosis: A 10-year study of 341 autogenous vein grafts

Purpose: Although duplex surveillance of infrainguinal bypass grafts is wid ely accepted, the optimal frequency and intensity of graft surveillance rem ains controversial. Earlier reports have suggested that grafts can be strat ified into high-risk and low-risk groups based on the presence or absence o f early graft flow disturbances. The purpose of this study was to provide l ong-term data in determining whether early graft flow disturbances detected by means of duplex scanning can predict the development of intrinsic vein graft stenosis. Methods: We reviewed a series of patients undergoing prospective duplex gra ft surveillance after autogenous infrainguinal bypass grafting procedures f rom 1987 to 1997. Patients included in the study underwent at least one dup lex scan within 3 months of graft implantation and were observed for a mini mum of 6 months. Grafts were categorized as abnormal when a focal flow dist urbance with a peak systolic velocity greater than 150 cm/s was identified within 3 months of graft implantation. Results: Of 341 vein grafts in 296 patients who met inclusion criteria, 89 grafts (26%) required revision for intrinsic stenosis; the mean follow-up p eriod was 35 months (range, 6 months to 10 years). Early flow disturbances were detected in 84 (25%) grafts. Grafts with early flow disturbances were more likely to ultimately require revision (43% vs 21%; P =.0001) and requi red initial revision earlier (8 months vs 16 months; P = .019). Eighty-two percent of initial graft revisions occurred in the first 2 postoperative ye ars; 69% occurred in the first year. However, an annual 2% to 4% incidence of late-appearing graft stenosis persisted during long-term follow-up. An a dditional 24 patients (7% of grafts) required an inflow or outflow reconstr uction. Conclusion: Grafts with early postoperative flow disturbances detected by m eans of duplex scanning have nearly three times the incidence of graft-thre atening stenosis and an earlier requirement for revision, when compared wit h normal grafts. This suggests that the biology and etiology of these lesio ns may differ. These data support not only aggressive efforts to detect ear ly graft lesions to stratify grafts at highest risk, but also continued lif elong graft-surveillance to detect late-appearing lesions, inflow and outfl ow disease progression, and maximize graft patency.