Abstract
Endoscopic vein harvesting (EVH) is the standard of care for coronary artery bypass grafting (CABG) in the United States, but recent comparisons with open harvesting suggest that conduit quality and outcomes may be compromised in EVH. To test the hypothesis that problems with EVH may relate to its learning curve and conduit quality, we analyzed the quality and early function of conduits procured by technicians with varying experience in EVH. Experienced (more than 900 cases, n = 55 patients) and novice (less than 100 cases, n = 30 patients) technicians performed EVH during CABG. Subsequently, optical coherence tomography (OCT) was used to examine the conduits for vascular injury, with segments identified as injured being further examined for gene expression with an array of genes related to tissue injury. Conduit diameter was measured intra- and postoperatively (day 5 and 6 months, respectively) with OCT and computed tomographic angiography. Endoscopic vein harvesting by novice harvesters resulted in a greater number of discrete graft injuries and greater expression of tissue-injury genes than EVH done by experienced harvesters. Regression analysis revealed an association between shear stress and early dilation of engrafted vessels (positive remodeling) (R 2 = 0.48, p < 0.01). Injured veins showed blunted positive remodeling at 5 days after harvesting and a greater degree of late lumen loss at 6 months. Under normal conditions, intraluminal shear stress leads to positive remodeling of vein grafts during the first postoperative week. Injury to conduits, a frequent sequela of the learning curve for EVH, was a predictor of early graft failure and of blunted positive remodeling and greater negative remodeling of endoscopically harvested vein grafts. Given the current annual volume of cases in which EVH is used, rigorous monitoring of the learning curve for this procedure represents an important and unrecognized issue in public health.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 11-18 |
| Number of pages | 8 |
| Journal | Annals of Thoracic Surgery |
| Volume | 93 |
| Issue number | 1 |
| DOIs | |
| State | Published - Jan 2012 |
Funding
Prior studies have demonstrated a variety of histopathologic abnormalities in SVG harvested through EVH [ 19 ], but have failed to address whether these abnormalities influence the performance of the harvested vessel as a bypass graft. Our study utilized high-resolution OCT and CTA imaging to identify graft injury within the grafted portion of SVG and to postoperatively follow serial changes in the diameter of the conduit created with the graft. The design of the study allowed a link to be established between severe vascular injury (ie, 4 or more discrete injuries) and abnormal positive remodeling of grafts over the first 5 days after CABG. Saphenous vein grafts spared from injury showed a strong correlation between the calculated intraluminal shear and the degree of graft dilation measured on day 5. In contrast, injured SVG showed little or no vasodilatory response to shear and showed a strong trend toward a greater risk of late lumen loss, or negative remodeling, at 6 months. Lumen loss of grafts within the first year after CABG is critical, because grafts with smaller luminal diameters are more prone to late graft failure [ 20 ]. Early positive remodeling occurs in SVG when nitric oxide, prostacyclins, and matrix metalloproteinases are produced locally in response to a chronic increase in intraluminal shear stress in the graft [ corroborating the importance of adventitial injury and a fibrotic scar response in the negative postoperative remodeling of grafts. 16, 21 ]. Animal and clinical studies demonstrate that positive remodeling requires intact endothelium [ 18, 22 ] and correlates with improved long-term SVG patency [ 13 ]. Impairment of this remodeling response has been attributed to trauma as well as to biochemical and morphologic changes within the graft [ 16 ]. To increase their elastic modulus in response to an increase in wall tension, SVG develop wall thickening and stiffening [ 13, 23 ]. However, narrowing of the SVG lumen within the first year after CABG is predominantly caused by negative remodeling of the entire vessel rather than by changes in wall thickness and stiffness [ 10 ]. Evidence suggests that adventitial injury and the subsequent fibrotic scar response at the site of injury are critical steps in the pathogenesis of negative remodeling of grafts after CABG [ 24–26 ]. We documented that the vascular and adventitial injury noted in our SVG were associated with the increased expression of genes linked to tissue injury and to a remodeling response ( Table 3 ), As the outermost layer of the vein, the adventitia is exposed to injury by EVH, particularly when novice technicians perform the initial blunt dissection step in this procedure. The ability to insure that the perivascular and adventitial tissues are fully preserved with EVH stands in stark contrast to the open, “no touch” technique of harvesting espoused by Souza and colleagues [ 11, 12 ]. Their technique describes the procurement of a full pedicle of fatty tissue completely encircling the vein without dissection of the vessel wall itself. With this as the gold standard for comparison, it is clear that the use of EVH to minimize graft injury so as to procure venous conduits with normal postoperative function requires that the technicians doing the harvesting acquire a higher degree of technical mastery than previously estimated. Indeed, we confirm that technician inexperience increased the risk of abnormalities in the quality and function of vascular conduits procured by EVH. As compared with those harvested by the expert group of PAs in our study, veins harvested by the novice group had significantly more tears identified in their intimal and deep vascular layers. Moreover, the positive remodeling response induced by shear was significantly blunted in veins procured by novices as compared with those procured by the experienced group of PAs in our study. We have demonstrated that the degree of injury sustained by a graft affects its function and early patency [ 27 ]. These findings support our hypothesis that the learning curve for EVH influences the quality and function of bypass conduits, and warrant further study to more fully characterize this learning curve. The strengths of our study were that unlike all prior analyses of the effect of EVH on conduit quality, it quantified injury within the portion of conduit that was chosen for grafting. Previous studies have used sampling of discarded graft material for histologic and microscopic analysis. However, such studies are limited in their ability to permit conclusions that can be generalized to the rest of the graft because they use only a small portion of vessel, and presumably one that is disproportionately poor in quality as compared with the rest of the graft. By comparison, we noted no differences in our two study groups as assessed with the standard learning-curve metrics of the time needed to complete EVH, the number of sutures required for repair of the harvested vein, and the need to convert from EVH to an open harvesting technique. Additionally, our analysis of messenger RNA expression provided evidence that the injury observed in SVG was associated with changes in the cellular biology of the vessel and was not simply an incidental finding. Our study was limited by its small cohort size and lack of use of conventional angiography for confirming our perioperative findings with OCT and CTA. Without this gold standard we consider the results of our work to be a proof of concept rather than an exhaustive validation of our study protocol. However, defining the natural history and clinical importance of early SVG injury has remained elusive, largely because of the inconvenience of performing serial invasive angiography. In view of the variety of clinical advantages of serial imaging with OCT and CTA, future clinical research that incorporates their use may provide a renewed opportunity to make progress with this type of analysis. In conclusion, intraluminal shear stress leads to positive remodeling of normal vein grafts during the first week after CABG. Poor conduit quality, a sequela of the learning curve for EVH, was a predictor of early graft failure, blunted positive remodeling, and greater negative remodeling of SVG after CABG. Given the large annual volume of use of EVH, rigorous monitoring of the learning curve for this procedure represents an important and under-recognized public health issue. This study was supported by National Institutes of Health Grant RO1 HL084080 , awarded to Robert S. Poston, MD.
ASJC Scopus subject areas
- Cardiology and Cardiovascular Medicine
- Pulmonary and Respiratory Medicine
- Surgery
