(Circulation. 1997;96:1707-1710.)
© 1997 American Heart Association, Inc.
Articles |
From The Cleveland (Ohio) Clinic Foundation, Department of Cardiology.
Correspondence to Stephen G. Ellis, MD, The Cleveland Clinic Foundation, 9500 Euclid Ave, F-25, Cleveland, OH 44195. E-mail elliss{at}cesmtp.ccf.org
Key Words: Editorials diabetes mellitus angioplasty bypass
| Introduction |
|---|
|
|
|---|
In aggregate, 4310 patients with multivessel disease thought to be suitable for either form of revascularization (thereby excluding many patients with far advanced disease) were enrolled in six RCTs between 1986 and 1991. The overall trial results were remarkably concordant. CABG was associated with a slight but not statistically significant survival advantage, less angina, and far fewer later revascularizations. PTCA led to a slight but insignificant reduction in myocardial infarction over the ensuing 2 to 5 years.1 2 3 4 5 6
Critics of RCTs often contest the generalizability of the treatment outcomes reported. They question whether it might be an oversimplification to apply the overall results of a trial both to all of its component patients and also to all similar but nonrandomized patients. In fact, given the general homogeneity engendered by the focus of most clinical trials, it is unusual for some patients to benefit and others to be harmed by the same treatment. Nonetheless, a well-designed RCT will a priori designate certain patient subsets to see whether similar overall treatment effects are observed across key subgroups. The number of these subgroups should be limited to minimize the risk of spurious identification of apparently significant findings (type I statistical error), as exemplified by the classic Gemini-Libra "effect" in the ISIS-2 trial.7
In the design of the BARI trial, four such subgroups of patients were specified. Patients were categorized on the basis of severity of angina, number of diseased vessels, ventricular function, and complexity of the lesions to be treated. In 1992, the BARI Data and Safety Monitoring Board requested an analysis of the treatment effect in diabetic patients because of findings in the TIMI 2 trial, which found diabetics to have a higher mortality when treated with PTCA within 18 to 48 hours of myocardial infarction than those randomized to a more conservative strategy.8 When the treatment effect on the 353 randomized BARI patients with treated diabetes was studied, there was a distinct survival advantage associated with early CABG, with the survival curves beginning to separate as early as 6 months after randomization and a gradual continued augmentation of this separation out to 5 years of follow-up (80.6% versus 65.5%, P=.003).6 Among randomizable but not randomized diabetics who tended to have less advanced disease, there was no survival difference.
In this issue of Circulation, the BARI investigators explore potential causes for this differential treatment effect in randomized diabetics and ascribe much of it to the benefit of the long-term patency of IMAs for reduction of fatality in the event of myocardial infarction.9
Given the origins of the BARI findings and acknowledging the statistically powerful treatment effect noted, one should still ask, is the finding that diabetics have a large survival advantage with bypass surgery reasonable? Does it have a pathophysiological basis, and did the other major RCTs have similar findings? One should further ask whether other patient subgroups might have a similar differential response to therapy and whether the IMA explanation is reasonable and scientifically founded.
Substantial evidence exists that diabetics respond differently than nondiabetics to both PTCA and CABG. After bypass surgery, diabetes is associated with more rapid progression of disease in both grafted and nongrafted native arteries and within SVGs.10 The influence of diabetes on IMA attrition has not been systematically studied. Diabetes is also a strong independent risk factor for late cardiovascular mortality after CABG,11 although data from the CASS registry suggest improved late survival in older diabetics randomized to surgical rather than medical therapy.12 Despite similar acute procedural success and complication rates, diabetics who undergo PTCA are at increased risk for restenosis and demonstrate significantly elevated rates of myocardial infarction, repeat revascularization, and overall mortality compared with nondiabetics in long-term follow-up.13
The adverse outcomes observed in diabetics after either form of revascularization appear to be a function of a variety of underlying pathophysiological abnormalities that accompany the disease.14 Metabolic and hematologic derangements, including dyslipidemia, enhanced platelet aggregability, and increased circulating levels of procoagulants, contribute to the elevated risk of subsequent myocardial infarction. When myocardial infarction does occur in diabetics, it is associated with an approximately twofold increase in acute mortality, reinfarction, and late mortality.15 Furthermore, diabetics, independent of the extent of underlying coronary atherosclerosis, are more prone to develop congestive heart failure (diabetic cardiomyopathy) than nondiabetics.
With this background, to put the BARI results into perspective, it is useful to appraise the results of other large RCTs. In CABRI, RITA, and EAST (233 diabetics total), the 5-year overall mortalities in the PTCA and CABG groups were 15% and 12%, respectively (M. Bertrand, MD, S. Pocock, Professor of Medical Statistics, and S. King, MD, 1997, personal communication).
IMA grafts, primarily as a result of their resistance to atherosclerosis, are associated with long-term patency rates far superior to those observed for SVGs. In large observational reports with up to 15 years of follow-up, this has translated into significant and progressive survival benefits for CABG patients who received an IMA graft as opposed to only an SVG.16 It has been theorized that the survival advantage afforded by the IMA graft results from the use of this vessel to provide prolonged protection for the typically extensive left anterior descending coronary artery territory. The enhanced survival associated with IMA grafting in these reports also probably reflects a bias toward use of the IMA in younger patients with less extensive coronary disease, less unstable angina, and less left ventricular dysfunction than patients who received only SVGs.
In the model devised by the BARI investigators to determine the reasons for the mortality benefit observed in diabetics who underwent CABG as opposed to PTCA, it appeared that the survival advantage was confined to the subset of surgical patients who received an IMA graft. Contrary to previous reports, patients who received an IMA graft were no less likely to experience a myocardial infarction during the follow-up period than patients who received only vein grafts. Surprisingly, the mortality benefit afforded by the presence of an IMA graft applied not only to patients who suffered a myocardial infarction during the follow-up period but also to diabetics (but not nondiabetics) without intercurrent infarction.
Why should IMA use (but not the use of vein grafts or PTCA) decrease mortality for diabetics without myocardial infarction? Perhaps this finding is simply a manifestation of shortcomings in the model for IMA use employed in this study. The decision to use an IMA graft was left to the discretion of the individual surgeon, permitting the potential for selection bias that could not be accounted for by the post hoc correction score used. Perhaps the true incidence of myocardial infarction was simply underestimated in diabetics, given their propensity for silent infarcts. Finally, the absolute number of patients on whom these subgroup comparisons were based is very small. Only 33 of the original 1829 patients (<2%) enrolled in BARI who underwent surgery and did not receive an IMA graft were diabetics. Conclusions based on small numbers of selected patients in retrospectively determined subgroups should be drawn with caution.
Although the BARI report9 provides practicable clinical guidelines for revascularization of diabetics with multivessel disease, it is not fully scientifically satisfying. It is difficult to assume that the only explanatory hypothesis that the BARI investigators tested was enhanced survival of patients receiving IMAs. One might reasonably expect that they assessed whether or not the benefit was confined to insulin-dependent diabetics or patients with three or more treatable lesions, to name but two of a multitude of possibilities. It is difficult to judge the validity of their conclusion that IMA use is pivotal without knowing how many other hypotheses were tested and rejected. Furthermore, their conclusions would be greatly enhanced if they were combined with the results of angiographic follow-up in these patients.
Søren Kierkegaard wrote, "life can only be understood backward, but it must be lived forward." The techniques of CABG and PTCA have undergone major changes since the 1980s, when these studies were performed. Use of multiarterial graft conduits is much more common and hospital stays shorter with CABG. Surgeons at many centers are exploring techniques that further hasten recovery. The long-term impact (especially graft patency) of these new revascularization strategies is unknown. Percutaneous revascularization may have evolved to an even greater extent. Metallic stents, shown to improve long-term results in several patient subsets,17 18 are now used in 40% to 50% of procedures in the United States and to an even greater extent overseas. Platelet glycoprotein IIb/IIIa inhibitors, demonstrated to dramatically reduce the early complications of angioplasty,19 are used in 30% to 35% of procedures in the United States. In fact, only a small minority of patients treated with percutaneous intervention in 1997 are treated as they would have been in BARI and other RCTs of that time period. Clearly, device and pharmacological approaches in the operating theater and in the catheterization laboratory will continue to evolve.
Given Kierkegaard's admonition, one should be circumspect about predicting the future, but to make the question raised by the BARI report on diabetics relevant, one must ask how diabetics might fare with the therapies of today and how they will fare with the therapies of tomorrow. For the results of percutaneous transluminal coronary revascularization to rival those of CABG, plaque rupture and thrombosis in coronary segments that could have been bypassed, restenosis with its attendant risk of treatment, and the acute complications of percutaneous revascularization must be minimized. With the exception of statin therapy, which benefits diabetics as well as nondiabetics,20 improved understanding of the molecular mechanisms of plaque rupture21 have not yet been translated into clinical benefit.
Restenosis after vascular injury results from a combination of mechanical forces (immediate vessel recoil), proliferative events (neointimal hyperplasia), and/or late changes in vessel geometry (remodeling). Coronary stenting, which essentially eliminates recoil and late constrictive remodeling, is associated with reduced restenosis in selected patients.17 18 However, neointimal hyperplasia, which is believed to be the process primarily responsible for in-stent restenosis when it does occur, also appears to be the predominant mechanism responsible for the heightened incidence of restenosis in diabetics.22 Although coronary stent placement may also reduce the incidence of restenosis in diabetic populations, preliminary observational reports suggest that restenosis after stent placement remains more frequent in diabetics than in nondiabetics. Data from two case series involving 163 diabetics and 876 nondiabetics indicate an elevated incidence of restenosis in the diabetic cohort (42.3% versus 22.9%).23 24 However, restenosis rates were significantly reduced by stent placement compared with PTCA in the 92 diabetics enrolled in the STRESS I and II trials (24% versus 60%, P<.01).25
Platelet glycoprotein IIb/IIIa receptor blockade is associated with a reduction in procedural complications and improved clinical outcome after PTCA, but the mechanism of long-term benefit remains uncertain. Given an increased number of glycoprotein IIb/IIIa receptors per platelet and increased baseline platelet aggregability in diabetics, the benefits of potent platelet inhibition should extend to this group of patients. However, experimental studies suggest that blood vessel passivation after vascular injury is delayed in diabetic animals.26 Preliminary subgroup analysis from the EPILOG study, which demonstrated a reduction in composite ischemic events 6 months after PTCA in patients randomized to abciximab rather than placebo (22.5% versus 25.8%, P=.023), found no treatment benefit in diabetics (27.4% versus 27.0%, P=.866).
Several emerging strategies that specifically target
neointimal formation after PTCA, if effective, might prove
especially helpful in the diabetic population. Directed pharmacological
blockade of the
vß3 receptor,
present on the surface of platelets and other cellular elements
in the blood vessel wall, has produced striking reductions in
neointimal hypoplasia in animal models, although clinical
studies have yet to be performed.27 Locally delivered
ionizing radiation, aimed at inhibiting smooth muscle cell
proliferation and migration after vascular injury, has resulted in the
attenuation of restenosis in animal models. Preliminary human
data suggest that
radiation decreases restenosis in
coronary arteries stented for treatment of
restenosis,28 and this therapy appears to be
particularly effective in diabetics (P. Teirstein, MD, personal
communication, 1997). Gene therapy, another novel approach, involves
either the administration of short DNA sequences (antisense
oligonucleotides) designed to inhibit the translation
of specific proto-oncogenes critical to smooth muscle cell
proliferation or, alternatively, the introduction of genetic material
in an attempt to augment production of desired cellular
products. Transfer of genes coding for nitric oxide and VEGF, two
of many promising compounds that may attenuate late vessel narrowing,
has recently been achieved in animal models.29 30 Although
it remains uncertain whether diabetics may be more tolerant or
resistant to the uptake or expression of genetic material than
nondiabetics, experimental data from the atherosclerotic rabbit model
indicate that efficient gene transfer is possible in extensively
diseased vessels.31 Nonetheless, just as the relative
importance of anatomic mechanisms of restenosis differs in
diabetics and nondiabetics (neointimal hyperplasia is
relatively more important in diabetics), it would not be surprising if
the relative importance of molecular mechanisms also differs.
Preliminary data suggest that wound-induced VEGF expression may be
diminished in diabetics compared with nondiabetics.32
Finally, clinical trials examining the efficacy of "tight"
diabetic control, preferably for at least 1 month before and extending
3 to 6 months after PTCA, are needed.
Management of diabetic cardiovascular disease will continue to pose a difficult challenge for physicians well into the next century. The BARI study findings, despite limitations, will help the clinician choose the best current therapy but, perhaps more importantly, will suggest questions the solutions to which will ultimately lead to better therapies.
| Selected Abbreviations and Acronyms |
|---|
|
| Footnotes |
|---|
| References |
|---|
|
|
|---|

ß3 integrin with
a small peptide antagonist GpenGRGDSPCA. J
Vasc Surg. 1994;19:125-134.[Medline]
[Order article via Infotrieve]This article has been cited by other articles:
![]() |
M. L. Brown, T. M. Sundt III, and B. J. Gersh Indications for Revascularization Card. Surg. Adult, January 1, 2008; 3(2008): 551 - 572. [Full Text] |
||||
![]() |
R. T. Hurst and R. W. Lee Increased Incidence of Coronary Atherosclerosis in Type 2 Diabetes Mellitus: Mechanisms and Management Ann Intern Med, November 18, 2003; 139(10): 824 - 834. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. S. Rumsfeld, D. J. Magid, M. E. Plomondon, J. Sacks, W. Henderson, M. Hlatky, G. Sethi, D. A. Morrison, and Veterans Affairs Angina With Extremely Serious Ope Health-related quality of life after percutaneous coronary intervention versus coronary bypass surgery in high-risk patients with medically refractory ischemia J. Am. Coll. Cardiol., May 21, 2003; 41(10): 1732 - 1738. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. K. McGuire, K. J. Anstrom, and E. D. Peterson Influence of the Bypass Angioplasty Revascularization Investigation National Heart, Lung, and Blood Institute Diabetic Clinical Alert on Practice Patterns: Results from the National Cardiovascular Network Database Circulation, April 15, 2003; 107(14): 1864 - 1870. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. M. Sundt III, B. J. Gersh, and H. C. Smith Indications for Coronary Revascularization Card. Surg. Adult, January 1, 2003; 2(2003): 541 - 559. [Full Text] |
||||
![]() |
J. H. Cole, E. L. Jones, J. M. Craver, R. A. Guyton, D. C. Morris, J. S. Douglas Jr, Z. Ghazzal, and W. S. Weintraub Outcomes of repeat revascularization in diabetic patients with prior coronary surgery J. Am. Coll. Cardiol., December 4, 2002; 40(11): 1968 - 1975. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Bigi, A. Desideri, L. Cortigiani, J. J. Bax, L. Celegon, and C. Fiorentini Stress Echocardiography for Risk Stratification of Diabetic Patients With Known or Suspected Coronary Artery Disease Diabetes Care, September 1, 2001; 24(9): 1596 - 1601. [Abstract] [Full Text] [PDF] |
||||
![]() |
S.-i. Yamagishi, D. Edelstein, X.-l. Du, and M. Brownlee Hyperglycemia Potentiates Collagen-Induced Platelet Activation Through Mitochondrial Superoxide Overproduction Diabetes, June 1, 2001; 50(6): 1491 - 1494. [Abstract] [Full Text] |
||||
![]() |
N. W. Niles, P. D. McGrath, D. Malenka, H. Quinton, D. Wennberg, S. J. Shubrooks, J. F. Tryzelaar, R. Clough, M. J. Hearne, F. Hernandez Jr, et al. Survival of patients with diabetes and multivessel coronary artery disease after surgical or percutaneous coronary revascularization: results of a large regional prospective study J. Am. Coll. Cardiol., March 15, 2001; 37(4): 1008 - 1015. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Van Belle, R. Ketelers, C. Bauters, M. Perie, K. Abolmaali, F. Richard, J.-M. Lablanche, E. P. McFadden, and M. E. Bertrand Patency of Percutaneous Transluminal Coronary Angioplasty Sites at 6-Month Angiographic Follow-Up : A Key Determinant of Survival in Diabetics After Coronary Balloon Angioplasty Circulation, March 6, 2001; 103(9): 1218 - 1224. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. M. Shindler, S. T. Palmeri, T. A. Antonelli, L. A. Sleeper, J. Boland, T. P. Cocke, J. S. Hochman, and for the SHOCK Investigators Diabetes mellitus in cardiogenic shock complicating acute myocardial infarction: a report from the SHOCK Trial Registry J. Am. Coll. Cardiol., September 1, 2000; 36(3_Suppl_A): 1097 - 1103. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. L. Bhatt, S. P. Marso, A. M. Lincoff, K. E. Wolski, S. G. Ellis, and E. J. Topol Abciximab reduces mortality in diabetics following percutaneous coronary intervention J. Am. Coll. Cardiol., March 15, 2000; 35(4): 922 - 928. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. A. Eagle, R. A. Guyton, R. Davidoff, G. A. Ewy, J. Fonger, T. J. Gardner, J. P. Gott, H. C. Herrmann, R. A. Marlow, W. C. Nugent, et al. ACC/AHA guidelines for coronary artery bypass graft surgery: A report of the American College of Cardiology/ American Heart Association task force on Practice Guidelines (Committee to revise the 1991 Guidelines for Coronary Artery Bypass Graft Surgery) J. Am. Coll. Cardiol., October 1, 1999; 34(4): 1262 - 1347. [Full Text] [PDF] |
||||
![]() |
R. E. Kuntz Importance of Considering Atherosclerosis Progression When Choosing a Coronary Revascularization Strategy : The Diabetes–Percutaneous Transluminal Coronary Angioplasty Dilemma Circulation, February 23, 1999; 99(7): 847 - 851. [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1997 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |