Abstract
Background
The internal mammary artery (IMA) is the preferred conduit for bypassing the left anterior descending (LAD) artery in patients undergoing coronary artery bypass grafting (CABG). Systematic evaluation of the frequency and predictors of IMA failure and long-term outcomes is lacking.
Methods and Results
PREVENT IV trial participants who underwent IMA-LAD revascularization and had 12–18-month angiographic follow-up (n=1539) were included. Logistic regression with fast false selection rate methods was used to identify characteristics associated with IMA failure (≥75% stenosis). The relationship between IMA failure and long-term outcomes including death, myocardial infarction, and repeat revascularization was assessed using Cox regression. IMA failure occurred in 132 participants (8.6%). Predictors of IMA graft failure were LAD stenosis <75% (odds ratio [OR], 1.76; 95% confidence interval [CI], 1.19–2.59), additional bypass graft to diagonal branch (OR, 1.92; 95% CI, 1.33–2.76), and not having diabetes (OR, 1.82; 95% CI, 1.20–2.78). LAD stenosis and additional diagonal graft remained predictive of IMA failure in an alternative model that included angiographic failure or death before angiography as the outcome. IMA failure was associated with a significantly higher incidence of subsequent acute (<14 days of angiography) clinical events, mostly due to a higher rate of repeat revascularization.
Conclusions
IMA failure was common and associated with higher rates of repeat revascularization, and patients with intermediate LAD stenosis or with an additional bypass graft to the diagonal branch had increased risk for IMA failure. These findings raise concerns regarding competitive flow and the benefit of CABG in intermediate LAD stenosis without functional evidence of ischemia.
Clinical Trial Registration Information
ClinicalTrials.gov. Identifier: NCT00042081.
Keywords: internal mammary artery, internal thoracic artery, graft failure, coronary artery bypass surgery
The internal mammary artery (IMA) conduit is considered the gold standard for bypassing the left anterior descending (LAD) coronary artery in patients undergoing coronary artery bypass grafting (CABG) surgery.1 The IMA-to-LAD graft has been shown to be more durable than other arterial and vein grafts as well as coronary stents for treatment of LAD disease, with patency rates >90% at 5-year follow-up.2–6 Additionally, most of the survival benefit observed in patients who undergo CABG compared with percutaneous coronary intervention (PCI) or medical management might be attributed to the patency of IMA-to-LAD.7,8 Despite the graft’s common use, high-quality studies of IMA graft failure are limited in number and small in sample size, lack systematic angiographic evaluation in all participants irrespective of symptoms, and/or have limited clinical follow-up.9,10 We therefore sought to identify and characterize factors associated with IMA graft failure and investigate the relationship between IMA graft failure and clinical outcomes in patients undergoing isolated CABG using data collected in the Project of Ex-vivo Vein Graft Engineering via Transfection (PREVENT) IV trial.11–12
Methods
Study Population
The design, primary results, and long-term follow-up of PREVENT IV have been published previously.11–13 In short, PREVENT IV was a double-blind, multicenter, randomized clinical trial in which edifoligide (an E2F transcription factor decoy) was compared with placebo in ex vivo treatment of vein grafts during CABG surgery. The trial enrolled 3014 participants at 107 U.S. sites between 2002 and 2003. The PREVENT IV protocol was approved by the institutional review boards of participating medical centers, and all patients gave informed consent. Participants were required to have primary, isolated CABG with at least 2 planned vein grafts, and without comorbid illness that would significantly limit life expectancy. The first 2400 participants enrolled in PREVENT IV were assigned to an angiographic cohort; of this group, 1829 patients returned for scheduled coronary angiography at 12 to 18 months after surgery. Per protocol, participants in the angiographic cohort who underwent angiography for clinical reasons and had graft failure before programmed angiographic follow-up were exempt from additional angiography. As shown in Figure 1, the starting population of the current analysis was the complete PREVENT IV trial population. From this cohort, we excluded 269 participants in whom an IMA graft was not used to revascularize the LAD, diagonal branches, or any graft combination that involved the LAD and/or diagonal branches. Participants (n=559) who were not enrolled in the angiographic cohort were also excluded. Of the remaining cohort of 2186 participants, 481 were excluded because of loss to follow-up and 79 because of death prior to angiographic follow-up. Of the remaining 1626 participants who underwent cardiac catheterization, 87 were excluded because of missing data on IMA patency. The remaining 1539 participants subsequently represented the study population for the main analysis.
Figure 1.
Flow diagram of study population. IMA-LAD indicates internal mammary artery-left anterior descending artery; CABG, coronary artery bypass surgery; f/u, follow-up.
Outcomes, Definitions, and Event Adjudication
The primary outcome of this study was IMA graft failure, defined as stenosis of at least 75% or occlusion of the diameter of the lumen of the IMA graft, which was assessed by quantitative coronary angiography 12 to 18 months after surgery, or earlier when performed as part of medical care and IMA graft failure was found. Since 79 participants died before angiography, we performed a sensitivity analysis in which IMA graft failure was defined as a composite of angiographic determined graft failure or death prior to angiography. The secondary outcomes were clinical outcomes among participants who did or did not have IMA graft failure; these included death and composites of 1) death or myocardial infarction (MI) or 2) death, MI, or repeat revascularization through 5 years from index CABG. All angiograms were analyzed at a core laboratory (PERFUSE Angiographic Core Laboratory, Boston, MA). Clinical events were assessed annually through mail and telephone contact with the participants. Five-year follow-up was complete in 95.1% of participants; consent for participation was withdrawn by 2.0% of participants, and an additional 2.9% were lost to follow-up. For reported events, medical records were collected and adjudicated by an independent clinical events committee using prespecified criteria.
Statistical Analysis
Baseline characteristics and medications among participants with and without IMA graft failure were summarized by medians and 25th and 75th percentiles for continuous variables and by percentages for categorical variables. Fast false selection rate (Fast FSR) in forward selection was used to select variables associated with graft failure in a logistic regression model. Fast FSR is a variable selection method that controls the rate that uninformative variables enter the model.14 Candidate variables were identified based on existing literature and investigator opinion, including age, sex, history of diabetes, renal failure, hypercholesterolemia, smoking, previous MI, ejection fraction, elective surgery, preoperative maximum stenosis of LAD, preoperative maximum stenosis of left main coronary artery, cardiopulmonary bypass, bypass graft quality, target artery quality, sequential versus single graft, and additional bypass graft to diagonal branch of LAD. Graft and target artery quality were graded as “good, fair, or poor” based on qualitative criteria by the surgeon at the time of surgery. Linearity assumptions were assessed for continuous variables and transformations applied when applicable. Odds ratios and 95% confidence intervals (CIs) were used to assess the relationship between the model variables and IMA graft failure.
Of those participants in the angiographic cohort who had an IMA graft, 481 were lost to follow-up and 79 died before angiography. Two sensitivity analyses were performed to address the biases introduced by competing risk and/or inability to perform angiographic follow-up. First, we assessed the composite endpoint of death or IMA graft failure. In the second, we used inverse probability weighting for missing data such that participants were weighted according to their estimated probability of having observed angiographic follow-up.15
Event rates for postangiography clinical outcomes in participants with and without IMA graft failure were calculated with Kaplan-Meier methods among participants event-free at angiographic follow-up. When the postangiography clinical outcomes were assessed, the day of the angiography was considered day 0. Hazard ratios (HRs) and 95% CIs were used to assess the relationship of per-patient IMA graft failure and clinical outcomes, and were calculated with the Cox regression model. Covariates adjusted for in the postangiography clinical outcome models included variables that were associated with IMA graft failure and death before angiography. There was a strong indication of nonproportional hazards in the relationship between IMA graft failure and the composite outcome of death, MI, or revascularization. Immediately after protocol-mandated angiography, the rate of revascularization was substantially higher than during longer follow-up. To account for this, we estimated different HRs for the periods before and after 14 days for outcomes containing revascularization.
Results
Baseline and Perioperative Characteristics
The baseline clinical, angiographic, and perioperative characteristics of the PREVENT IV total and angiographic populations with use of IMA-LAD graft are displayed in Table 1. Of the IMA grafts included in this analysis, 99.4% were left IMA grafts. Baseline and surgical characteristics tended to be similar in the angiographic cohort with IMA use compared with the overall PREVENT IV trial population who received an IMA graft (n=2745). Among the angiographic population with IMA use (n=1539), a total of 132 (8.6%) had IMA graft failure at follow-up (angiogram: 350–563 days after randomization). Among those with IMA graft failure, 68 had an IMA graft stenosis of 75–95%, 3 had a subtotal stenosis (95–99%), and 61 had an occluded IMA graft. In general, participants who had IMA graft failure less frequently had diabetes mellitus, had less severe stenosis of the LAD, and had similar target artery quality but worse graft quality as assessed by the operator during surgery. Other clinical and procedural characteristics were similar. Participants with IMA graft failure had similar surgery duration, and in-hospital recovery, as measured by intensive care and hospital length of stay.
Table 1.
Clinical, Angiographic, and Procedural Characteristics
| Total Cohort With IMA-LAD Graft (n=2745) | Angiographic Population With IMA-LAD Grafts (n=1539) | Angiographic Population With IMA-LAD Graft Failure (n=132) | Angiographic Population Without IMA-LAD Graft Failure (n=1406) | |
|---|---|---|---|---|
| Preoperative characteristics | ||||
| Age,* y | 63 (56, 70) | 63 (55, 69) | 63 (56, 68) | 63 (55, 69) |
| Male, % | 80.1 | 82.1 | 77.3 | 82.5 |
| BMI,* kg/m2 | 29 (26, 33) | 29 (26, 42) | 29 (26, 33) | 28 (26, 31) |
| Weight,* kg | 88 (77, 100) | 88 (78, 100) | 84 (73, 93) | 89 (79, 100) |
| Race, % | ||||
| White | 91.0 | 90.8 | 85.6 | 91.3 |
| Black | 4.4 | 4.6 | 12.1 | 3.8 |
| Other | 4.6 | 4.6 | 2.3 | 4.8 |
| Current smoker, % | 22.3 | 20.9 | 25.0 | 20.5 |
| Diabetes mellitus, % | 38.0 | 36.0 | 25.0 | 37.1 |
| Hypercholesterolemia, % | 76.9 | 77.9 | 73.5 | 78.3 |
| Hypertension, % | 74.7 | 72.8 | 66.7 | 73.4 |
| Congestive heart failure, % | 9.5 | 6.6 | 5.3 | 6.8 |
| Renal failure, % | 2.2 | 1.1 | 1.5 | 1.1 |
| Cerebrovascular disease, % | 12.3 | 8.6 | 5.3 | 8.9 |
| Peripheral vascular disease, % | 11.6 | 10.1 | 6.8 | 10.4 |
| Chronic lung disease, % | 15.0 | 12.9 | 14.4 | 12.8 |
| Previous MI, % | 41.9 | 42.1 | 47.7 | 41.5 |
| Previous PCI, % | 25.4 | 27.0 | 29.5 | 26.8 |
| History of cancer, % | 7.9 | 7.5 | 11.4 | 7.1 |
| Ejection fraction,* % | 50 (40, 60) | 50 (40, 60) | 50 (45, 60) | 50 (40, 60) |
| Angiographic characteristics | ||||
| LAD stenosis, % | ||||
| <50% | 5.3 | 5.1 | 8.4 | 4.9 |
| 50–74% | 18.6 | 17.6 | 24.4 | 16.9 |
| 75–95% | 48.6 | 47.7 | 40.5 | 48.4 |
| 95–99% | 17.5 | 18.5 | 21.4 | 18.2 |
| 100% | 10.1 | 11.1 | 5.3 | 11.7 |
| LM stenosis, % | ||||
| 0% | 40.0 | 43.5 | 45.8 | 43.2 |
| <50% | 33.3 | 29.3 | 30.5 | 29.2 |
| 50–74% | 17.3 | 17.6 | 15.3 | 17.8 |
| 75–95% | 7.9 | 8.3 | 8.4 | 8.3 |
| 95–99% | 1.3 | 1.2 | 0.0 | 1.4 |
| 100% | 0.1 | 0.2 | 0.0 | 0.2 |
| No. diseased vessels* | 2 (2, 3) | 2 (2, 3) | 2 (2, 3) | 2 (1, 3) |
| Procedural characteristics | ||||
| Elective procedure, % | — | 51.1 | 47.3 | 51.4 |
| Cardiopulmonary bypass, % | 78.3 | 78.5 | 75.8 | 78.7 |
| Left IMA use, % | 99.4 | 99.3 | 98.5 | 99.4 |
| Free IMA graft, % | 1.4 | 1.2 | 2.6 | 1.1 |
| Sequential IMA graft, % | 5.4 | 5.3 | 7.6 | 5.0 |
| Graft quality, % | ||||
| Good | 86.4 | 91.3 | 86.5 | 91.8 |
| Fair | 7.9 | 8.1 | 11.1 | 7.8 |
| Poor | 0.7 | 0.6 | 2.4 | 0.5 |
| Target artery quality, % | ||||
| Good | 67.7 | 70.0 | 74.4 | 69.6 |
| Fair | 22.5 | 22.0 | 18.6 | 22.3 |
| Poor | 8.0 | 8.0 | 7.0 | 8.1 |
| Surgery duration,* min | 232 (194, 272) | 227 (191, 270) | 227 (191, 270) | 231 (192, 274) |
| CPB duration,* min | 100 (80, 123) | 99 (80, 122) | 99 (80, 121) | 102 (81, 123) |
| Time on ventilator,* h | 7 (5, 13) | 7 (5, 12) | 7 (4, 12) | 8 (5, 14) |
| ICU length of stay,* h | 26 (22, 47) | 25 (21, 44) | 25 (21, 44) | 27 (22, 48) |
| Hospital length of stay,* days | 6 (5, 8) | 6 (5, 7) | 6 (5, 7) | 6 (5, 8) |
IMA-LAD indicates internal mammary artery-left anterior descending artery; BMI, body mass index; MI, myocardial infarction; PCI, percutaneous coronary intervention; LM, left main coronary artery; CPB, cardiopulmonary bypass; ICU, intensive care unit.
Median (IQR).
Medications at 1 Year
At 1 year, similar proportions of patients with and without IMA graft failure were taking aspirin (95.3% and 94.4%, respectively; P=0.68), beta-blockers (76.6% and 73.2%, respectively; P=0.41), nitrates (8.6% and 5.6%, respectively; P=0.16), and statins (82.8% and 80.9%, respectively; P=0.59). More patients with than without IMA failure were taking clopidogrel or ticlopidine (33.6% and 20.8%, respectively; P=0.0008), and fewer patients with than without IMA failure were taking an angiotensin-converting enzyme inhibitor or angiotensin II receptor blocker (44.5% and 57.2%, respectively; P=0.006).
Predictors for IMA Graft Failure
As shown in Table 2, the absence of diabetes, less severe preoperative LAD stenosis, and an additional bypass graft to the diagonal were associated with IMA graft failure at follow-up. Using an alternative approach in which the outcome included IMA graft failure or death prior to angiography, preoperative LAD stenosis and having an additional bypass graft to the diagonal remained associated with IMA graft failure, but not having diabetes did not. A sensitivity analysis using inverse-probability-weighted methods yielded similar results (data displayed as Supplemental Table 1).
Table 2.
Predictors for IMA-LAD Graft Failure
| OR | 95% CI | P | |
|---|---|---|---|
| IMA graft failure* | |||
| Preoperative LAD stenosis (<75%) | 1.86 | 1.26–2.75 | 0.002 |
| Not having diabetes | 1.86 | 1.22–2.81 | 0.004 |
| Additional bypass graft to diagonal | 1.92 | 1.33–2.76 | 0.001 |
| IMA graft failure or death prior to cath† | |||
| Age, per year | 1.03 | 1.01–1.05 | 0.001 |
| Female sex | 1.55 | 1.09–2.20 | 0.015 |
| Hypercholesterolemia | 0.67 | 0.48–0.94 | 0.019 |
| Renal failure | 3.57 | 1.45–8.84 | 0.006 |
| Preoperative LAD stenosis (<75%) | 1.49 | 1.07–2.08 | 0.017 |
| Current smoker | 1.48 | 1.02–2.13 | 0.038 |
| Additional bypass graft to diagonal | 1.49 | 1.11–2.01 | 0.009 |
IMA-LAD indicates internal mammary artery-left anterior descending artery.
IMA graft failure model was built on data from 1396 participants with complete data on all candidate variables and was fit on 1538 participants with complete data on the chosen variables.
The predictive model for the composite of IMA graft failure or death prior to cath was built on 1467 participants with complete data on all candidate variables and was fit on 1616 participants with complete data on the chosen variables.
Events Between CABG Surgery and Angiographic Follow-Up
Of the 132 participants with IMA graft failure, 4 had an MI between CABG surgery and angiographic follow-up, and 9 had a repeat revascularization procedure; of these procedures, 5 were due to IMA graft failure. Of the 1407 participants without IMA graft failure, MI occurred in 14 and repeat revascularization in 45 between CABG surgery and angiographic follow-up.
IMA Graft Failure and Clinical Outcomes
Duration of follow-up was 1815±245 days in participants with IMA failure and 1813±239 days in those without IMA failure. The occurrence of clinical events from follow-up angiography among participants with and without IMA graft failure is illustrated in Figure 2. As shown, a markedly higher rate of the composite of death, MI, or repeat revascularization was found within the first 14 days after angiography among participants with IMA graft failure compared with those without IMA graft failure (14.4% vs. 4.9%, adjusted HR, 3.92; 95% CI, 2.30–6.68, p<0.0001). After the first 14 days, a trend toward more events was observed after IMA graft failure over 4 years of follow-up (adjusted HR, 1.45; 95% CI, 0.85–2.48, p=0.17). This association was predominantly driven by an association between IMA graft failure and repeat revascularization, with less of a relationship between IMA graft failure and either death or the composite of death or MI (Table 3). When assessing outcomes by IMA graft stenosis severity (Table 4), we found that participants with total IMA graft occlusion were more likely to have a subsequent cardiovascular event (45.9% vs. 36.8%), including death (8.2% vs. 2.9%), compared with participants with non-occluded, failed IMA grafts.
Figure 2.
Relation between internal mammary artery (IMA) graft to left anterior descending artery failure and the composite of death, myocardial infarction, or repeat revascularization over 4 years after angiographic follow-up.
Table 3.
Relationship of IMA Graft Failure With Clinical Outcomes
| Outcome | No IMA Graft | IMA Graft | HR | 95% CI | P |
|---|---|---|---|---|---|
| Failure n (%) | Failure n (%) | ||||
| Death, MI, or revasc | |||||
| ≤ 14 days | 84 (5.5) | 19 (14.4) | 3.92 | 2.30–6.68 | <0.0001 |
| >14 days | 172 (11.8) | 17 (15.0) | 1.45 | 0.85–2.48 | 0.17 |
| Death or MI | 118 (7.7) | 14 (10.6) | 1.29 | 0.66–2.50 | 0.46 |
| Death | 83 (5.4) | 7 (5.3) | 1.10 | 0.50–2.39 | 0.82 |
IMA indicates internal mammary artery; MI, myocardial infarction. Outcomes were adjusted for age, gender, hypercholesterolemia, renal failure, preoperative left anterior descending artery stenosis, current smoker, and diabetes. These models were fit on 1355, 1400, and 1480 participants, respectively, who had complete data on the variables adjusted for.
Table 4.
Number of Clinical Events Among Participants With IMA Graft Failure (≥75% Stenosis) by LAD Stenosis Severity
| IMA Graft Stenosis at Follow-up
|
||||
|---|---|---|---|---|
| <75% (n=1407) | 75–95% (n=68) | 95–99% (n=3) | 100% (n=61) | |
| Death, MI, or repeat revasc, n (%) | 271 (19.3%) | 25 (36.8%) | 1 (33.3%) | 28 (45.9%) |
| Death or MI, n (%) | 104 (7.4%) | 6 (8.8%) | 0 | 8 (13.1%) |
| Death, n (%) | 76 (5.4%) | 2 (2.9%) | 0 | 5 (8.2%) |
| MI, n (%) | 30 (2.1%) | 4 (5.9%) | 0 | 3 (4.9%) |
| Repeat revasc, n (%) | 195 (13.9%) | 21 (30.9%) | 1 (33.3%) | 23 (37.7%) |
IMA indicates internal mammary artery; MI, myocardial infarction.
Discussion
In the current study, we sought to identify characteristics associated with IMA graft failure and determine its relationship with subsequent clinical outcomes. We found that patients with less severe preoperative LAD stenosis and those who had an additional bypass graft placed to the diagonal branch were at higher risk for IMA graft failure. When restricting the analysis to those who completed angiographic follow-up, the absence of diabetes was also associated with IMA graft failure. IMA graft failure led to increased risk for early adverse clinical events, mostly repeat revascularization, while the association with long-term clinical outcomes was less clear. To our knowledge, this analysis represents the first robust assessment of IMA graft failure and long-term clinical outcomes in a large cohort of patients undergoing CABG surgery with systematic angiographic follow-up. Evidence from previous studies on predictors of IMA graft failure is conflicting with respect to the relation between IMA graft failure and the severity of the stenosis in native vessels. Angiographic predictors for IMA graft failure were assessed by Manninen et al16 and by Gaudino et al17 over several years of follow-up, and the severity of stenosis of the native vessel was not found to influence the patency rate of the IMA bypass graft.
In contrast, 2 other small studies have suggested that arterial graft failure was associated with low-grade recipient artery stenosis and competitive flow.18,19 The study with the most clinical impact was performed by Berger et al,9 in which the authors assessed the long-term patency of IMA bypass grafts in 273 grafts in 230 patients. IMA graft failure was defined as diffuse and >95% conduit narrowing (“string sign”), and angiographic follow-up was performed for clinical reasons. Using stepwise multivariable logistic regression, the authors found that lower diameter stenosis of the native artery was the only significant predictor for IMA graft failure (odds ratio, 21.5; 95% CI, 5.18–64.38). This observation was also found when the analysis was restricted to IMA-LAD grafts. Despite the differences in the definitions for IMA graft failure, the findings of this study are strikingly similar to the results of our study and imply that the decision to use an IMA should be carefully considered in light of the severity of the stenosis of the target vessel. In addition, our study shows that one should also take into account the risk of introducing “competitive flow” when placing an additional bypass graft to the diagonal branch of the LAD. Our study clearly delineates the negative impact on the patency of the IMA-LAD graft, as illustrated by an almost 2 times higher occurrence of graft failure.
Adaptation of the IMA Graft to the Coronary Circulation
The benefit of IMA grafts over vein grafts has been consistently observed irrespective of age, sex, stenosis severity, or left ventricular function, with widening survival differences over time.7,20,21 A number of risk factors have been identified to result in IMA graft failure, including hypercholesterolemia, lipoproteins, diabetes mellitus, smoking, age, and prior chest radiation (which may negatively influence the quality of the arterial graft). There are a number of reasons why IMA graft failure may be less common than for other grafts. During harvesting and suturing, the IMA graft is more resilient after surgical trauma, due to its muscular wall and abundant collateral blood supply, and endothelial and vascular wall function are therefore better preserved compared with vein grafts.22 Also, the diameter of the IMA graft is generally closely matched with the diameter of the LAD, which limits flow turbulences that create frictional force and shear stress-induced endothelial injury. However, these differences do not fully explain the superior patency and resistance to atherosclerosis of the IMA graft compared with vein grafts and coronary arteries. A number of structural and physical properties of the IMA graft are additionally thought to play a role.23 The endothelium of the IMA may play a key role, as it is markedly different from other bypass grafts. Specifically, it has fewer fenestrations and lower intracellular junction permeability, which may prevent lipoproteins from entering the subendothelial space.24 In addition, the endothelial cells of the IMA are rich in heparin sulfate and endothelial nitric oxide synthase (eNOS). When eNOS is expressed, it produces low concentrations of nitric oxide, which preserves good endothelial function and integrity, allows arterial remodeling in response to flow (enlargement of arterial lumen, or intimal thickening for reduction in vessel lumen), and contributes to the antithrombotic properties and endothelial homeostasis, which in turn confer protection from atherosclerosis.24,25
Mechanisms of IMA Failure
When IMA graft failure occurs, technical error is the most common cause in the early postoperative period. In the subsequent weeks to months, localized neointimal hyperplasia may occur at the cleft between the native artery and the IMA graft at the anastomotic suture site, on the hood and on the floor of the native LAD, which can result in a localized stenosis.24 Performing an IMA bypass on a low-grade LAD stenosis or supplying an infarcted LAD territory leads to competitive flow and low flow, respectively, and adversely affects IMA graft patency (“disuse atrophy”). Causes for late (and rare) IMA failure include progressive fibrointimal proliferation and/or atherosclerosis either in the IMA graft or in the native LAD vessel.23 In this study, we assessed IMA graft patency at 12–18 months. Unfortunately, we know little about when the graft actually failed.
Need for Physiologic Assessment in Angiographic Intermediate Lesions
Our study raises concerns regarding the performance of CABG with the use of IMA in the treatment of native vessels with only mild or moderate stenosis. Although competitive flow between native and grafted coronary vessels is likely at play here, the use of angiography as a surrogate for physiologic impact of these intermediate severe lesions is imprecise.26 The measurement of hyperemic translesional pressure ratio, or fractional flow reserve (FFR), has been applied to study coronary bypass graft failure and has shown that lower graft occlusion rates are found in participants with FFR-assessed functionally significant lesions compared with functionally non-significant lesions (n=164, 9% vs. 21% at 1 year).27 The angiographic percentage diameter stenosis in this study displayed a similar but less precise correlation with graft failure. While unnecessary grafting of functionally nonsignificant lesions is believed to be of little clinical consequence, it may affect the decision to perform CABG instead of PCI when it involves the LAD in the setting of multivessel disease.
Limitations
The present study has a number of limitations. First, we measured IMA graft failure at 12–18 months after CABG, which excluded cases in which failure would have led to a fatal outcome prior to angiography (i.e., those with diabetes mellitus). We tried to account for this bias by including early deaths in the endpoint for IMA graft failure. Second, the timing of IMA graft failure before angiography was unknown, and whether this coincided with nonfatal events that occurred between CABG surgery and angiographic follow-up is unknown. Third, physiologic assessment was not performed in angiographic intermediate lesions, which would have provided the ability to discriminate whether a lesion was flow limiting and whether competitive flow was a cause for IMA graft failure. Fourth, protocol-driven angiography may have influenced revascularization after angiography and subsequent clinical events. Fifth, missing data due to loss of follow-up might also have affected our results. Lastly, the study may have been underpowered to detect a modest but clinically important relationship between IMA graft failure and subsequent long-term clinical outcomes.
Conclusions
We found that IMA graft failure was common and that bypassing intermediate-grade LAD stenosis or introducing competitive flow by placing an additional bypass graft to the diagonal branch significantly increased the risk of subsequent IMA graft failure. IMA graft failure at 12–18 months was associated with higher rates of early repeat revascularization but not long-term subsequent clinical outcomes. The decision to perform CABG with the use of an IMA graft should be made carefully when coronary angiography indicates an intermediate LAD lesion, and the use of functional assessment of ischemia may be warranted.
Supplementary Material
Clinical Perspectives.
In this study, the frequency, predictors, and impact of internal mammary artery (IMA) graft failure were evaluated in patients undergoing coronary artery bypass grafting (CABG). Using data from the PREVENT-IV trial, we identified 1539 participants with IMA-left anterior descending (LAD) graft and protocol angiography at 12–18 months after the index procedure. Our main findings included that IMA graft failure (defined as ≥75% stenosis) occurred in 132 participants (8.6%) at angiographic follow-up. Independent predictors of IMA graft failure were low-grade LAD stenosis and not having diabetes. LAD stenosis, but not diabetes, remained predictive of IMA graft failure in an alternative model that included IMA failure or death before angiography as the outcome. IMA failure was associated with a significantly higher incidence of subsequent acute (within 14 days of angiography) clinical events, mostly due to a higher rate of repeat revascularization. This study represents the first robust assessment of IMA graft failure and long-term clinical outcomes in a large cohort of patients undergoing CABG surgery with systematic angiographic follow-up, irrespective of symptom status. Our study raises concerns regarding the performance of CABG with the use of IMA in the treatment of native vessels with only mild or moderate stenosis. As such, it confirms findings from smaller studies which have suggested that the severity of LAD stenosis is of key importance for patency of the IMA-LAD graft, which is thought to be responsible for the survival benefit observed in clinical studies that compared CABG with multivessel percutaneous coronary intervention and/or medical therapy.
Acknowledgments
Funding Sources: Ms Hager’s statistical work was funded by an NIH training grant (T32HL079896). Drs Alexander and Smith are supported in part by grant U01-HL088953 from the NIH Cardiothoracic Surgical Trials Network.
Footnotes
Disclosures: Disclosures for Drs Alexander, Peterson, and Lopes are available at https://dcri.org/about-us/conflict-of-interest. Dr. Harrington has received research grants and consultant/advisory fees from Merck & Co.
The other authors report no potential conflicts of interest.
References
- 1.Hillis LD, Smith PK, Anderson JL, Bittl JA, Bridges CR, Byrne JG, Cigarroa JE, Disesa VJ, Hiratzka LF, Hutter AM, Jr, Jessen ME, Keeley EC, Lahey SJ, Lange RA, London MJ, Mack MJ, Patel MR, Puskas JD, Sabik JF, Selnes O, Shahian DM, Trost JC, Winniford MD. 2011 ACCF/AHA Guideline for Coronary Artery Bypass Graft Surgery: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. Circulation. 2011;124:e652–735. doi: 10.1161/CIR.0b013e31823c074e. [DOI] [PubMed] [Google Scholar]
- 2.Khot UN, Friedman DT, Pettersson G, Smedira NG, Li J, Ellis SG. Radial artery bypass grafts have an increased occurrence of angiographically severe stenosis and occlusion compared with left internal mammary arteries and saphenous vein grafts. Circulation. 2004;109:2086–2091. doi: 10.1161/01.CIR.0000127570.20508.5C. [DOI] [PubMed] [Google Scholar]
- 3.Harskamp RE, Lopes RD, Baisden CE, de Winter RJ, Alexander JH. Saphenous vein graft failure after coronary artery bypass surgery: pathophysiology, management, and future directions. Ann Surg. 2013;257:824–833. doi: 10.1097/SLA.0b013e318288c38d. [DOI] [PubMed] [Google Scholar]
- 4.Etienne PY, D'Hoore W, Papadatos S, Mairy Y, El Khoury G, Noirhomme P, Hanet C, Glineur D. Five-year follow-up of drug-eluting stents implantation vs minimally invasive direct coronary artery bypass for left anterior descending artery disease: a propensity score analysis. Eur J Cardiothoracic Surg. 2013;44:884–890. doi: 10.1093/ejcts/ezt137. [DOI] [PubMed] [Google Scholar]
- 5.Goldman S, Zadina K, Moritz T, Ovitt T, Sethi G, Copeland JG, Thottapurathu L, Krasnicka B, Ellis N, Anderson RJ, Henderson W. Long-term patency of saphenous vein and left internal mammary artery grafts after coronary artery bypass surgery: results from a Department of Veterans Affairs Cooperative Study. J Am Coll Cardiol. 2004;44:2149–2156. doi: 10.1016/j.jacc.2004.08.064. [DOI] [PubMed] [Google Scholar]
- 6.Boylan MJ, Lytle BW, Loop FD, Taylor PC, Borsh JA, Goormastic M, Cosgrove DM. Surgical treatment of isolated left anterior descending coronary stenosis. Comparison of left internal mammary artery and venous autograft at 18 to 20 years of follow-up. J Thorac Cardiovasc Surg. 1994;107:657–662. [PubMed] [Google Scholar]
- 7.Loop FD, Lytle BW, Cosgrove DM, Stewart RW, Goormastic M, Williams GW, Golding LA, Gill CC, Taylor PC, Sheldon WC, Proudfit WL. Influence of the internal-mammary-artery graft on 10-year survival and other cardiac events. N Engl J Med. 1986;314:1–6. doi: 10.1056/NEJM198601023140101. [DOI] [PubMed] [Google Scholar]
- 8.Hlatky MA, Boothroyd DB, Bravata DM, Boersma E, Booth J, Brooks MM, Carrié D, Clayton TC, Danchin N, Flather M, Hamm CW, Hueb WA, Kähler J, Kelsey SF, King SB, Kosinski AS, Lopes N, McDonald KM, Rodriguez A, Serruys P, Sigwart U, Stables RH, Owens DK, Pocock SJ. Coronary artery bypass surgery compared with percutaneous coronary interventions for multivessel disease: a collaborative analysis of individual patient data from ten randomised trials. Lancet. 2009;373:1190–1197. doi: 10.1016/S0140-6736(09)60552-3. [DOI] [PubMed] [Google Scholar]
- 9.Berger A, MacCarthy PA, Siebert U, Carlier S, Wijns W, Heyndrickx G, Bartunek J, Vanermen H, De Bruyne B. Long-term patency of internal mammary artery bypass grafts: relationship with preoperative severity of the native coronary artery stenosis. Circulation. 2004;110:II36–II40. doi: 10.1161/01.CIR.0000141256.05740.69. [DOI] [PubMed] [Google Scholar]
- 10.Cameron A, Davis KB, Green GE, Myers WO, Pettinger M. Clinical implications of internal mammary artery bypass grafts: the Coronary Artery Surgery Study experience. Circulation. 1988;77:815–819. doi: 10.1161/01.cir.77.4.815. [DOI] [PubMed] [Google Scholar]
- 11.Alexander JH, Ferguson TB, Jr, Joseph DM, Mack MJ, Wolf RK, Gibson CM, Gennevois D, Lorenz TJ, Harrington RA, Peterson ED, Lee KL, Califf RM, Kouchoukos NT. The PRoject of Ex-vivo Vein graft ENgineering via Transfection IV (PREVENT IV) trial: study rationale, design, and baseline patient characteristics. Am Heart J. 2005;150:643–649. doi: 10.1016/j.ahj.2005.05.021. [DOI] [PubMed] [Google Scholar]
- 12.Alexander JH, Hafley G, Harrington RA, Peterson ED, Ferguson TB, Jr, Lorenz TJ, Goyal A, Gibson M, Mack MJ, Gennevois D, Califf RM, Kouchoukos NT. Efficacy and safety of edifoligide, an E2F transcription factor decoy, for prevention of vein graft failure following coronary artery bypass graft surgery: PREVENT IV: a randomized controlled trial. JAMA. 2005;294:2446–2454. doi: 10.1001/jama.294.19.2446. [DOI] [PubMed] [Google Scholar]
- 13.Lopes RD, Williams JB, Mehta RH, Reyes EM, Hafley GE, Allen KB, Mack MJ, Peterson ED, Harrington RA, Gibson CM, Califf RM, Kouchoukos NT, Ferguson TB, Lorenz TJ, Alexander JH. Edifoligide and long-term outcomes after coronary artery bypass grafting: PRoject of Ex-vivo Vein graft ENgineering via Transfection IV (PREVENT IV) 5-year results. Am Heart J. 2012;164:379–386. e1. doi: 10.1016/j.ahj.2012.05.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Boos DD, Stefanski LA, Wu Y. Fast FSR variable selection with applications to clinical trials. Biometrics. 2009;65:692–700. doi: 10.1111/j.1541-0420.2008.01127.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Seaman SR, White IR. Review of inverse probability weighting for dealing with missing data. Stat Methods Med Res. 2013;22:278–295. doi: 10.1177/0962280210395740. [DOI] [PubMed] [Google Scholar]
- 16.Manninen HI, Jaakkola P, Suhonen M, Rehnberg S, Vuorenniemi R, Matsi PJ. Angiographic predictors of graft patency and disease progression after coronary artery bypass grafting with arterial and venous grafts. Ann Thorac Surg. 1998;66:1289–1294. doi: 10.1016/s0003-4975(98)00757-7. [DOI] [PubMed] [Google Scholar]
- 17.Gaudino M, Alessandrini F, Nasso G, Bruno P, Manzoli A, Possati G. Severity of coronary artery stenosis at preoperative angiography and midterm mammary graft status. Ann Thorac Surg. 2002;74:119–121. doi: 10.1016/s0003-4975(02)03614-7. [DOI] [PubMed] [Google Scholar]
- 18.Hashimoto H, Isshiki T, Ikari Y, Hara K, Saeki F, Tamura T, Yamaguchi T, Suma H. Effects of competitive blood flow on arterial graft patency and diameter. Medium-term postoperative follow-up. J Thorac Cardiovasc Surg. 1996;111:399–407. doi: 10.1016/s0022-5223(96)70449-x. [DOI] [PubMed] [Google Scholar]
- 19.Villareal RP, Mathur VS. The string phenomenon: an important cause of internal mammary artery graft failure. Tex Heart Inst J. 2000;27:346–349. [PMC free article] [PubMed] [Google Scholar]
- 20.Loop FD. Internal-thoracic-artery grafts. Biologically better coronary arteries. N Engl J Med. 1996;334:263–265. doi: 10.1056/NEJM199601253340411. [DOI] [PubMed] [Google Scholar]
- 21.Hlatky MA, Boothroyd DB, Reitz BA, Shilane DA, Baker LC, Go AS. Adoption and effectiveness of internal mammary artery grafting in coronary artery bypass surgery among medicare beneficiaries. J Am Coll Cardiol. 2014;63:33–39. doi: 10.1016/j.jacc.2013.08.1632. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Kay HR, Korns ME, Flemma RJ, Tector AJ, Lepley D., Jr Atherosclerosis of the internal mammary artery. Ann Thorac Surg. 1976;21:504–507. doi: 10.1016/s0003-4975(10)63917-3. [DOI] [PubMed] [Google Scholar]
- 23.Shelton ME, Forman MB, Virmani R, Bajaj A, Stoney WS, Atkinson JB. A comparison of morphologic and angiographic findings in long-term internal mammary artery and saphenous vein bypass grafts. J Am Coll Cardiol. 1988;11:297–307. doi: 10.1016/0735-1097(88)90094-0. [DOI] [PubMed] [Google Scholar]
- 24.Otsuka F, Yahagi K, Sakakura K, Virmani R. Why is the mammary artery so special and what protects it from atherosclerosis? Ann Cardiothorac Surg. 2013;2:519–526. doi: 10.3978/j.issn.2225-319X.2013.07.06. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.He GW, Fan L, Grove KL, Furnary A, Yang Q. Expression and function of endothelial nitric oxide synthase messenger RNA and protein are higher in internal mammary than in radial arteries. Ann Thorac Surg. 2011;92:845–850. doi: 10.1016/j.athoracsur.2011.04.063. [DOI] [PubMed] [Google Scholar]
- 26.Sabik JF, 3rd, Lytle BW, Blackstone EH, Khan M, Houghtaling PL, Cosgrove DM. Does competitive flow reduce internal thoracic artery graft patency? Ann Thorac Surg. 2003;76:1490–1496. doi: 10.1016/s0003-4975(03)01022-1. [DOI] [PubMed] [Google Scholar]
- 27.Botman CJ, Schonberger J, Koolen S, Penn O, Botman H, Dib N, Eeckhout E, Pijls N. Does stenosis severity of native vessels influence bypass graft patency? A prospective fractional flow reserve-guided study. Ann Thoracic Surg. 2007;83:2093–2097. doi: 10.1016/j.athoracsur.2007.01.027. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.