Author + information
- ↵⁎Reprint requests and correspondence:
Dr. Giora Landesberg, Hebrew University, Hadassah Medical Center, Department of Anesthesiology and CCM, Kyriat-Hadassah, Ein Kerem, Jerusalem 91120, Israel
Exactly 30 years ago, Hertzer et al. (1) published the first in a series of reports on thousands of patients who underwent routine coronary angiography before major vascular surgery. Their landmark studies were the basis for our current recognition that vascular surgery patients have a high prevalence of significant coronary artery disease (CAD): 60% have 1 or more coronary arteries with >70% stenosis, 18% have severe triple-vessel disease, and 4% have left main disease (2). Subsequently, Hertzer et al. (1) showed that selective pre-operative coronary artery bypass graft surgery (CABG) in patients with severe CAD lowered the perioperative and long-term mortality relative to patients with similar degrees of CAD not treated with CABG. Numerous subsequent studies confirmed that major vascular surgery is associated with a high risk for both perioperative and long-term cardiac morbidity and mortality. Many other studies demonstrated that this risk is predicted by ischemia on pre-operative noninvasive cardiac testing (radionuclide imaging or dobutamine stress echocardiography). However, the data that pre-operative coronary revascularization (PCR), mainly by CABG, improves perioperative or long-term outcome after major vascular surgery came from retrospective observational studies only (3). Moreover, since PCR is not free of complications, serious questions were raised as to the overall risk-benefit ratio of PCR before major vascular surgery, questions that could only be answered by large-scale randomized controlled trials (RCTs).
Lately, 2 RCTs—the CARP (Coronary Artery Revascularization Prophylaxis) trial (4) and the DECREASE (Dutch Echocardiographic Cardiac Risk Evaluation Applying Stress Echo Study Group)-V (5) study—examined the role of PCR before major vascular surgery. Both trials found absolutely no perioperative or long-term benefit to PCR over medical therapy before vascular surgery. Unfortunately however, even these RCTs were not free of serious limitations. The CARP trial screened 5,859 vascular surgery patients from 18 Veterans Administration hospitals in the U.S., yet without a unified method of screening in all institutions. Subsequently, only a small proportion (8.9%) of screened patients was randomized. Only 44% of randomized patients had moderate-large ischemia on pre-operative nuclear imaging, and only 32% had triple-vessel disease (2.9% of all screened patients), indicating a potential selection bias with the possibility that mainly patients less likely to benefit from PCR were included in the trial. Patients with left-main CAD were excluded by design from randomization. Moreover, on reanalysis of the CARP study data, patients who had complete coronary revascularization with CABG had fewer post-operative cardiac complications (6), and among patients who were not included in the randomization, those with left main disease had better survival if treated with PCR (7). In contrast, the DECREASE-V trial aimed to randomize patients with the most severe CAD. However, this relatively small trial (n = 101) eventually randomized patients who were too ill for PCR and surgery. In the revascularization arm, 100% had history of myocardial infarction, 51% had ongoing angina, 47% had congestive heart failure, 37% had diabetes mellitus, 41% had a history of cerebrovascular accident, and 18% had renal failure, in addition to extensive ischemia on noninvasive pre-operative testing. Moreover, apparently all 49 patients randomly assigned to the revascularization arm were compelled to undergo PCR, although most studies show that up to one-third of vascular surgery patients with severe ischemia on noninvasive testing cannot undergo coronary revascularization because of unsuitable coronary anatomy, poor runoff, multiple small-vessel disease, or chronic total obstructions not amenable to percutaneous coronary intervention, or because they are too sick for CABG. Subsequently, the DECREASE-V trial patients suffered exceptionally high perioperative (11% to 22%) and 1-year mortality regardless of PCR. The authors of this editorial have previously shown that major vascular surgery patients with numerous risk factors have a poor post-operative prognosis irrespective of PCR (8). Therefore, major vascular surgery patients deserve careful pre-operative clinical assessment and judgment as to who may or may not benefit from coronary intervention and strict, blinded randomization can do harm to such patients.
Despite the above limitations, the impact of the CARP and DECEASE-V trials on clinical practice and published guidelines worldwide was enormous. Coronary revascularization is subsequently rarely recommended before major vascular surgery, and even the demand for noninvasive pre-operative cardiac testing in major vascular patients has decreased substantially. It is against this background that the study by Monaco et al. (9) in this issue of the Journalhas surprisingly emerged.
Monaco et al. (9) elegantly showed in a RCT that a strategy of routine pre-operative coronary angiography and subsequent selective PCR provides better long-term survival and event-free survival for patients undergoing abdominal aortic surgery, compared with a strategy of selective coronary angiography and PCR, performed only after pre-operative noninvasive testing showing significant ischemia. There was also a tendency to improved perioperative outcome (although not statistically significant). Included in the randomization were all patients with ≥2 RCRI criteria. Patients in the routine angiography arm underwent more PCRs (58.1% vs. 40.1%; p = 0.01). Interestingly, the beneficial effect of PCR in the routine angiography arm was on top of strong beta-blockade with the nonselective beta-blocker carvedilol, titrated to quite high doses to reach the effect of resting pre-operative heart rate of ≤60 beats/min. There are 2 main findings in this trial: 1) PCR improves outcome for this subset of medium-high risk vascular surgery patients; and 2) routine pre-operative coronary angiography provides better screening of vascular surgery patients than does noninvasive testing.
How can we reconcile the sharp conflict between the positive results of the trial by Monaco et al. (9) and the negative results of the previous 2 RCTs? First, unlike the previous RCTs, this trial included only patients undergoing abdominal aortic surgery and not lower extremity bypass operations. The latter are less stressful operations, associated with lower perioperative morbidity and mortality. Clearly, the DECREASE-V patients were much sicker, with more cardiac and noncardiac comorbidities than the patients in the present study with poorer prognosis regardless of any pre-operative treatment. Conversely, the patients in the present trial had more extensive CAD, including left main disease and a higher prevalence of triple-vessel disease than the CARP trial patients. Importantly, patients in the systematic pre-operative coronary angiography group underwent more PCRs with off-pump CABG than PCI compared with the control group (47.5% vs. 28.6%, respectively, with a tendency for statistical significance: p = 0.08). It is possible, therefore, that more patients with left main disease and triple-vessel disease underwent off-pump CABG than PCI, and these results translated into a significantly better long-term outcome in the systematic coronary angiography group who underwent more CABG than PCI, in corroboration with previous studies that found better outcome with CABG than PCI in similar patients (6,10).
Will this new RCT lead to a change in practice back to more pre-operative coronary revascularizations or to routine coronary angiography as Hertzer advocated 30 years ago? Time will tell. The study by Monaco et al. (9) suggests that routine pre-operative coronary angiography is better than noninvasive cardiac testing for detecting patients who may benefit from successful PCR. This relatively small trial must be corroborated by additional, preferably larger studies. The possibility of screening patients by computed tomography or magnetic resonance coronary angiography, rather than routine coronary angiography, should also be explored. The current practice of endovascular abdominal aneurysm repair, which is less stressful and associated with fewer perioperative complications, diminishes, although does not eliminate, the need for open aortic repairs, pre-operative cardiac evaluations, and eventual PCR. Nevertheless, it is safe to say, on the basis of currently accumulated data, that vascular surgery patients, particularly those with extensive and complex CAD and a high SYNTAX (Synergy between PCI with TAXUS and Cardiac Surgery) score (11), fare better in the long run with CABG than with PCI (6,10). Moreover, the current guidelines demanding prolonged dual-antiplatelet therapy, even after successful coronary stent implantation (at least 4 weeks for bare-metal stent and up to 1 year after drug-eluting stent implantation), limit the ability to use PCI as the method for coronary revascularization before major surgery. Above all, risk stratification of patients who are candidates for vascular surgery is frequently puzzling, and the treating physicians must rely on their best medical judgment whether to pursue pre-operative noninvasive or invasive testing as well as coronary revascularization procedures with these high-risk patients.
↵⁎ Editorials published in the Journal of the American College of Cardiologyreflect the views of the author and do not necessarily represent the view of JACCor the American College of Cardiology.
- American College of Cardiology Foundation
- Poldermans D.,
- Schouten O.,
- Vidakovic R.,
- et al.,
- DECREASE Study Group
- Monaco M.,
- Stassano P.,
- Di Tomasso L.,
- et al.
- Landesberg G.,
- Mosseri M.,
- Wolf Y.G.,
- et al.