HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): James E. Lowe Walter G. Wolfe H. Newland Oldham, Jr Peter Van Trigt, III Donald D. Glower Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Gall, S. Articles by Glower, D. D. Search for Related Content PubMed PubMed Citation Articles by Gall, S., Jr Articles by Glower, D. D.

Ann Thorac Surg 2000;69:524-530
© 2000 The Society of Thoracic Surgeons

---

Original Articles

Efficacy of the internal mammary artery in combined aortic valve replacement-coronary artery bypass grafting

^a Department of Surgery, Duke University Medical Center, Durham, North Carolina, USA

Address reprint requests to Dr Gall, Prairie Thoracic and Cardiovascular Surgeons 619 East Mason St, Springfield, IL 62701
e-mail: sagall{at}ibm.net

	Abstract

Top Abstract Introduction Patients and methods Results Comment References

 
Background. While internal mammary artery (IMA) use predicts improved survival after coronary bypass grafting (CABG), it remains unknown whether patients undergoing concomitant aortic valve replacement (AVR) realize a similar benefit.

Methods. All patients at a single teaching institution, undergoing combined AVR-CABG, which included a graft to the left anterior descending coronary artery (LAD) from 1984 to 1994 (n = 227) were examined retrospectively.

Results. Patients receiving an IMA graft (yesIMA, n = 135) and patients receiving only saphenous vein grafts (nonIMA, n = 92) were not different in their presenting symptoms, or in their incidence of preoperative risk factors. The patients with IMA were more likely to be male, have a later year of operation, be younger, and have a greater body surface. Morbidity was not different between groups. IMA use did not affect 30-day mortality. Long-term actuarial survival was greater in the group with IMA (63% ± 7% vs 42% ± 6% at 5 years, p< 0.01). A multivariate Cox proportional hazards model demonstrated that use of an IMA graft improved survival, while recent myocardial infarction, diabetes, earlier year of operation, and lower ejection fraction diminished long-term survival. The relative risk of IMA grafting was 0.570.

Conclusions. Within the limits of a retrospective analysis, patients in a modern era of cardiac operation, who undergo combined AVR-CABG, do not suffer increased morbidity from IMA use, and may realize a survival benefit from use of the IMA as a conduit for bypass of the LAD coronary artery.

	Introduction

Top Abstract Introduction Patients and methods Results Comment References

 
The results of aortic valve replacement (AVR) are affected by presence of coronary atherosclerosis. The initial report of orthotopic aortic valve replacement by Starr in 1963 [1], included coronary artery disease as a factor in the perioperative death of 1 patient. A limited understanding of the interaction of coronary artery disease and aortic valve disease resulted in initial acceptance of AVR as sole therapy for these patients. Reports involving small groups of patients undergoing AVR, noted that coronary atherosclerosis had a negative effect on hospital and long–term survival [2–4]. A larger group of patients was studied, to confirm that the presence of coronary atherosclerosis had an untoward effect on long–term survival in patients undergoing aortic valve replacement [5]. Anderson demonstrated that the combination of the individually effective coronary artery bypass grafting (CABG) and AVR therapies could be accomplished with an acceptable mortality [6]. Additional evaluation of combined procedures noted no increase in hospital mortality, or perioperative myocardial infarction, when careful intraoperative myocardial protection was provided [7].

While the life expectancy of patients with both aortic valve disease and coronary atherosclerosis is considered to be lower than that of isolated valve replacement, regardless of valvular disease [8], concomitant CABG negates the detrimental long–term effect of atherosclerotic coronary artery stenosis on survival in patients undergoing AVR [9]. Despite this finding, the effect of distribution and severity of CAD in AVR patients is not well studied. The use of the IMA as a bypass graft in isolated CABG was demonstrated to have an additional benefit for long–term survival [10]. Although alluded to by Lytle [11], it has not been clear whether use of the IMA as a bypass graft in combined AVR–CABG, where life expectancy is more limited than either disease in isolation, would result in improved survival or freedom from adverse ischemic events. We undertook this investigation to explore the hypothesis that IMA grafting would have a salutary effect on hospital and long–term (total) survival after combined AVR–CABG, and that this advantage would not be offset by excess in morbidity.

	Patients and methods

Top Abstract Introduction Patients and methods Results Comment References

 
All patients undergoing a cardiac valvular procedure at Duke University Medical Center from 1976 to 1994 were entered into a computerized database. Information was extracted retrospectively from the medical record, operative reports, and perfusion database. Patients were contacted, and medical records reviewed for long–term evaluation of signs and symptoms of coronary and valvular disease, as well as the end–points of repeat operation, other revascularization procedures, or death. Follow–up was concluded in July 1995.

A myocardial infarction was considered to be remote if it occurred more than 6 months prior to operation, and considered recent if it occurred less than 6 months prior to operation. Ejection fraction measurements were taken from left ventriculograms. Ventricular function, assessed by either method, which was graded simply as normal was assigned a missing value (n = 7 in final group), and that which was graded simply as impaired was assigned a missing value (n = 0). Coronary arterial stenoses of 75% or more were considered to be significant. Aortic valve gradient, area, and severity of regurgitation were calculated preferentially from catheterization data. Missing values from cardiac catheterization were supplied by values from echocardiographic evaluation where possible.

Mechanical ventilation for greater than 48 hours postoperatively constituted respiratory failure. Mediastinal exploration in the immediate postoperative period was considered as reoperation for bleeding. Reexploration of the sternal incision for reasons other than a mechanical complication, bleeding, or tamponade defined sternal wound infection. Low cardiac output syndrome was defined as cardiac index remaining less than 2.0 L · min^–1 · m^–2 for greater than or equal to 4.0 hours postoperatively. Inotrope use was instituted to achieve cardiac index greater than 2.0 L · min^–1 · m^–2, and usually was initiated with dopamine. The presence of new Q–waves on electrocardiograms was required for the diagnosis of perioperative myocardial infarction. Any patient who died within 30 days of operation, or in the hospital prior to discharge, was considered to have had a hospital death.

AVR was accomplished in 708 patients, of which 425 underwent combined AVR–CABG. The patients undergoing reoperative sternotomy and reoperative AVR were included. As a first step in providing a more homogenous population, all patients with prior CABG or undergoing other concomitant cardiac procedures were excluded. The IMA was used as a bypass conduit in this combined operation beginning in 1984, and anastomosed to the left anterior descending coronary artery (LAD) in all cases. Beginning in 1984, a total of 227 patients received, at a minimum, coronary bypass grafting of the LAD. These patients were analyzed in 2 groups, those receiving an IMA (yesIMA, n = 135) and those receiving a saphenous vein as a conduit to the LAD (nonIMA, n = 92). The remaining grafts in both groups were of saphenous vein. An additional group (nonLAD, n = 64) was identified as those patients undergoing AVR and CABG, but without grafting to the LAD. These patients were compared to the first 2 groups demographically, and evaluated for differences in long–term survival. Hypothermic, hyperkalemic cardioplegic cardiac arrest was used routinely for myocardial protection.

Statistical analysis was performed with SAS (SAS Institute, Cary, NC). Nonparametric Wilcoxon tests and ² analysis were used to compare variables between groups. Further differences between the 3 groups were assessed using Tukey’s test. Predictors of survival were analyzed using only those patients receiving LAD grafting. Variables determined preoperatively, which included demographic and constitutional factors, were analyzed separately from those variables determined during the conduct of the procedure. IMA use was included with the preoperative variables for its effect on survival. Factors that were significantly associated with hospital survival in univariate testing at the p less than or equal to 0.2 level were included in a multivariate logistic regression model. Long–term (total) survival analysis was performed by the product–limit method of Kaplan–Meier. Variables that were associated with long–term survival, at a level of the p less than or equal to 0.2, were included in a multivariate Cox proportional hazards model. Statistical significance for all analyses was taken to occur at a level of p less than 0.05.

	Results

Top Abstract Introduction Patients and methods Results Comment References

 
A total of 291 patients underwent AVR–CABG during the period of IMA use. Contained within this group, 64 patients (without LAD) did not undergo grafting of the LAD and 227 patients with LAD grafting were available for comparison of IMA (yesIMA, n = 135) versus saphenous vein only grafting (nonIMA, n = 92). Demographic and preoperative variables in the yesIMA and noIMA groups are compared in Table 1. Internal mammary grafts were first used at this institution for combined AVR–CABG in 1984, and initially accounted for 11% of patients who received a graft to the LAD. IMA use gradually increased such that by 1994, 75% of patients received an IMA as their conduit to the LAD. The yesIMA group had a later year of operation, and was also more likely to be male, be younger, and have a greater body surface area (BSA). BSA did not vary with year of operation. These 2 groups were not different in the severity of presenting symptoms or signs (Table 1).

Use of an IMA graft did not lengthen the time required on cardiopulmonary bypass or for aortic cross–clamp in patients who received a LAD graft (Table 4). The frequency of grafting of the right coronary artery and left circumferential coronary arteries and total number of grafts was not different between these 2 groups. The group, nonLAD, did have a significantly shorter period of cardiopulmonary bypass and aortic cross–clamp, while having fewer bypass grafts constructed. The average size of aortic prosthesis inserted in the yesIMA group was slightly larger than either the group nonLAD or nonIMA, but was no different when normalized to BSA (Table 4). The use of mechanical (n = 119) versus bioprosthetic (n = 108) valves was significantly different between the yesIMA and nonIMA groups. The yesIMA group received mechanical valves in 68% of cases, while the nonIMA patients received bioprosthetic valves in 71% of cases (p < 0.001). Bioprostheses were more likely than mechanical valves to be inserted in patients older than 70 years, than in those younger than 70 years (60% vs 39.5%, p < 0.01), an expected result. Valve sizes of the bioprosthetic and mechanical group were not different (22.6 + 2.1 vs 22.8 + 2.6 mm, p = not significant [NS]).

Hospital survival
There were 5 intraoperative deaths in the nonIMA group, and none in the yesIMA or nonLAD groups (p = 0.015). There was no difference between groups in the incidence of hospital death (nonIMA,11 vs yesIMA, 9 vs nonLAD, 2, p = NS). The causes of death in the hospital were considered to be cardiac in 16 patients and noncardiac in 6 patients. Logistic regression analysis identified the preoperative and perioperative variables predictive of hospital mortality (Table 5). The variables sex, ejection fraction, valve area, and valve area index made no contribution to predicted hospital mortality. Patients with perioperative myocardial infarction had a 100% mortality, and these events (n = 3) occurred exclusively in the yesIMA group. All of these myocardial infarctions occurred in the anterior wall of the left ventricle. Neither IMA use, nor underlying valvular condition, influenced overall hospital survival.

View larger version (16K): [in this window] [in a new window]   Fig 1. Actuarial survival following AVR-CABG comparing patients with an IMA graft to the LAD (yesIMA), and those with saphenous vein grafting of the LAD (nonIMA) (p = 0.0017).

View larger version (19K): [in this window] [in a new window]   Fig 2. Actuarial survival following AVR-CABG, comparing patients grouped by IMA use for LAD grafting and age greater than 70 years (overall, p = 0.0026; age > 70 vs age < 70, p = 0.0037).

	Comment

Top Abstract Introduction Patients and methods Results Comment References

 
This study was a retrospective analysis of the effects of IMA use on survival after combined AVR and CABG. The critical importance of grafting important coronary artery stenoses in patients undergoing aortic valve operation has been previously demonstrated [2, 9, 11–14]. Although a dissenting study did not support this conclusion, several aspects, including small sample size, a 10 year age difference between patients with and without coronary disease, limited use of cardioplegia, a short mean follow-up time, and the lack of a control group who had concomitant CABG limit its applicability [15]. Modern cardioplegic techniques allow adequate time for CABG and AVR to be safely accomplished [7, 11]. IMA use did not lengthen the ischemic period in this study, and may offer a particular advantage in that no proximal anastomosis is generally required for this graft.

Because no randomized evaluation of IMA use can be reasonably carried out, all patients operated upon in a single institution for concomitant AVR and CABG were included in this retrospective analysis. The period of evaluation was narrowed to include only those patients who were operated upon during the era when IMA grafting was available, to ensure comparison of contemporaneous patients. IMA grafting was not exclusively utilized during the majority of years of the study, and no significant differences exist between the nonIMA and yesIMA groups that may bias IMA use except age, sex, and BSA. BSA carries a small coefficient for its effect on hospital survival, and does not impact long-term survival. Although age did affect hospital survival, neither age nor sex was predictive of long-term survival in multivariate analysis. With these statistically significant, but biologically small differences in mind, this population remains reasonable to consider for analysis. Because all patients who received an IMA graft had the LAD grafted, the main comparison group of nonIMA patients was limited to include only those patients who received a minimum of a graft to the LAD. There were 10 patients (yesIMA, 4, nonIMA, 6) who received LAD grafting without a retrospectively identifiable LAD and/or left main stenosis greater than 75%. We relied on the surgeon’s judgment at the time of operation that a LAD graft was important, and therefore these patients remained in our analysis. By limiting the patient population to those receiving LAD grafting, 2 groups of patients with reasonably similar demographics and disease severity who are allocated to either treatment with LIMA, or in a control saphenous vein group and operated upon with similar techniques, can be compared in a meaningful way, despite the retrospective nature of this study.

Hospital mortality in this study decreased in later years of operation. The decade of this study encompassed changes in anesthetic and operative techniques, which were the likely cause of the significant decrease in hospital mortality (19.5% vs 3.3%) between the earlier and later halves of this study. Despite younger age and later year of operation in the yesIMA group, IMA use did not impact on hospital survival, and improvements in hospital survival over time occurred in both groups equally. The intraoperative factor, which negatively influenced hospital survival most strongly, was the occurrence of a perioperative myocardial infarction. Perioperative myocardial infarction was associated with hospital death in 3 patients. All of these infarctions were in the anterior distribution, and occurred in patients with IMA grafts. One patient ultimately died of small bowel infarction, while the other 2 died 3 days postoperatively of heart failure. No further information, concerning graft flow intraoperatively or time course of ischemia postoperatively, is available with regards to these patients. No other patients had any suggestion of an anterior hypoperfusion syndrome. Inadequate myocardial protection or technically inadequate anastomoses were the factors most likely to have caused such effects. IMA grafting is more demanding than the use of vein grafts, and care must be taken to ensure initial graft patency. Even with these deaths included in the yesIMA group, IMA use or nonuse did not influence hospital survival.

When total survival after AVR-CABG is examined, several preoperative factors independently influenced long-term survival (Table 6). While diabetes mellitus, operative year, and ejection fraction cannot be easily modified, postponement of an elective procedure following myocardial infarction and use of an IMA for LAD grafting are modifiable factors that can improve long-term survival. The benefit of delay after a recent myocardial infarction is suggested by the absence of an adverse effect from remote myocardial infarction. Deleterious effects from remote myocardial infarction are likely contained within the variable ejection fraction.

When intraoperative factors are considered, only a shorter duration of cardiopulmonary bypass and absence of myocardial infarction were independently important to long-term survival. Cardiopulmonary bypass time may be a measurable surrogate for factors of technical difficulty that extend the time of the procedure and thus the deleterious effects of cardiopulmonary bypass on end-organ function. This factor has a strong p-value, but a small coefficient and low relative risk, indicating significant risk only at more extreme values. The uniformly poor outcome in patients with perioperative myocardial infarction, although occurring in relatively few patients, reemphasizes the critical importance of intraoperative evaluation of graft flow and excellent myocardial preservation. This study was not able to examine the effects of myocardial injury less severe than transmural infarction.

Morbidity following AVR-CABG was uniformly low. Total operative time is a known independent predictor of wound infection. While total length of the operative procedure was not examined in this analysis, and IMA harvesting may add slightly to the total length of the procedure, no increased wound infection rate was noted in the yesIMA group. While IMA use has been postulated to increase pulmonary dysfunction, the yesIMA group actually suffered less pulmonary morbidity than the nonIMA group, although no data on the incidence or severity of pleural effusions was available in this population. Morbidity such as cerebrovascular events, cerebrovascular events with residual deficits, and respiratory failure were less likely in the yesIMA group, although this may have been a result of their slightly younger age [16]. The frequency of these complications did not change over the period of this study, while IMA use increased. While it may not be possible to evaluate other factors impacting morbidity, the increased use of IMA grafts directly correlated with decreased long-term mortality and did not result in any increase in morbidity after AVR-CABG.

The patients with coronary artery disease are not a homogenous group, and as noted in this study, differ widely in extent and distribution of disease as well as involvement of the LAD. The nonLAD group had less extensive coronary disease and slightly more severe valve disease, as well as having a larger proportion of women. Despite these slight differences in demographics and expected smaller burden of coronary disease, the nonLAD group was comparable to the groups with and without IMA. Examining the entire population of AVR-CABG patients operated upon during the IMA use era, the group of patients who required a LAD graft had a significantly diminished survival when compared to the nonLAD group. Those patients who received an IMA graft achieved survival similar to patients with no LAD grafting despite the advantages of fewer grafts and absence of LAD disease in the nonLAD group. Therefore, the use of an IMA conduit for LAD grafting neutralized the negative effect of LAD involvement such that these patients’ survival was equivalent to the nonLAD group (Fig 3).

View larger version (16K): [in this window] [in a new window]   Fig 3. Actuarial survival following AVR-CABG comparing patients without LAD grafting to those undergoing IMA or saphenous vein grafting to the LAD (nonLAD vs yesIMA, p = 0.227; nonIMA vs nonLAD, p = 0.0003; nonIMA vs yesIMA, p = 0.0017).

Use of the IMA has previously been demonstrated to improve survival in patients undergoing isolated CABG. This report demonstrates that, within the limitations of retrospective analysis, use of the IMA as a graft to the LAD in combined AVR-CABG has beneficial effects on long-term survival. IMA grafting of the LAD improves long-term survival in comparison to patients who received saphenous vein grafts only, effectively neutralizing the effect of important atherosclerotic involvement of the LAD. Coronary artery disease involving the LAD circulation should not be a factor in considering valve prosthesis selection when an IMA graft is available. Perioperative myocardial infarction is a rare, albeit important complication, portending a poor survival that mandates the utmost care in technical detail. Despite this infrequent, potentially lethal complication, use of the IMA resulted in no greater morbidity following operation. These benefits, and the absence of increased risk in using IMA grafts, support uniform use of IMA grafts in patients undergoing AVR-CABG with particular benefit in patients with LAD disease.

	References

Top Abstract Introduction Patients and methods Results Comment References

 

1. Starr A., Edwards M.L., McCord C.W., Griswold H.E. Aortic replacement. Circulation 1963;28:779-783.
2. Flemma R.J., Johnson W.D., Lepley D., Auer J.E., Tector A.J., Blitz J. Simultaneous valve replacement and aorta-to-coronary saphenous vein bypass. Ann Thorac Surg 1971;12:163-170.[Abstract/Free Full Text]
3. Linhart J.W., Wheat M.W. Myocardial dysfunction following aortic valve replacement. J Thorac Cardiovasc Surg 1967;54:259-269.
4. Peterson C.R., Herr R., Crisera R.V., Starr A., Bristow J.D., Griswold H.E. The failure of hemodynamic improvement after valve replacement surgery. Etiology, diagnosis, and treatment. Ann Int Med 1967;66:1-24.
5. Copeland J.G., Griepp R.B., Stinson E.B., Shumway N.E. Long-term follow-up after isolated aortic valve replacement. J Thorac Cardiovasc Surg 1977;74:875-889.[Abstract]
6. Anderson R.P., Bonchek L.I., Wood J.A., Chapman R.P., Starr A. The safety of combined aortic valve replacement and coronary bypass grafting. Ann Thorac Surg 1973;15:249-255.[Abstract/Free Full Text]
7. Kouchoukous N.T., Lell W.A., Rogers W.J. Combined aortic valve replacement and myocardial revascularization. Ann Surg 1983;197:721-727.[Medline]
8. Kirklin J.W., Barratt-Boyes B.G. Aortic valve disease. In: Kirklin J.W., Baratt-Boyes B.G., eds. Cardiac surgery. New York: Churchill Livingstone Inc, 1993:492-571.
9. Mullany C.J., Elveback L.R., Frye R.L., et al. Coronary artery disease and its management. J Am Coll Cardiol 1987;10:66-72.[Abstract]
10. Loop F.D., Lytle B.W., Cosgrove D.M., et al. Influence of the internal-mammary-artery graft on 10-year survival and other cardiac events. N Engl J Med 1986;314:1-6.[Medline]
11. Lytle B.W., Cosgrove D.M., Gill C.C., et al. Aortic valve replacement combined with myocardial revascularization. J Thorac Cardiovasc Surg 1988;95:402-414.[Abstract]
12. Nunley D.L., Grunkemeier G.L., Starr A. Aortic valve replacement with coronary bypass grafting. J Thorac Cardiovasc Surg 1983;85:705-711.[Abstract]
13. Czer L.S.C., Gray R.J., Steward M.E., Robertis M.D., Chaux A., Matloff J.M. Reduction in sudden late death by concomitant revascularization with aortic valve replacement. J Thorac Cardiovasc Surg 1988;95:390-401.[Abstract]
14. Miller D.C., Stinson E.B., Oyer P.E., et al. Surgical implications and results of combined aortic valve replacement and myocardial revascularization. Am J Cardiol 1979;43:494-501.[Medline]
15. Bonow R.O., Kent K.M., Rosing D.R., et al. Aortic valve replacement without myocardial revascularization in patients with combined aortic valvular and coronary artery disease. Circulation 1981;63:243-251.[Abstract/Free Full Text]
16. Roach G.W., Kanchuger M., Mangano C.M., et al. Adverse cerebral outcomes after coronary bypass surgery. N Engl J Med 1996;335:1857-1863.[Medline]

This article has been cited by other articles:

				N. D. Desai and G. T. Christakis Bioprosthetic Aortic Valve Replacement: Stented Pericardial and Porcine Valves Card. Surg. Adult, January 1, 2008; 3(2008): 857 - 894. [Full Text]

				J. M. Jones, D. Lovell, G. W. Cran, and S. W. MacGowan Impact of coronary artery bypass grafting on survival after aortic valve replacement Interact CardioVasc Thorac Surg, June 1, 2006; 5(3): 327 - 330. [Abstract] [Full Text] [PDF]

				S. Karthik, A. K. Srinivasan, A. D. Grayson, T. Friede, and B. M. Fabri Effect of the Left Internal Mammary Artery to the Left Anterior Descending Artery on Mortality and Morbidity After Combined Coronary and Valve Operations Ann. Thorac. Surg., July 1, 2005; 80(1): 163 - 169. [Abstract] [Full Text] [PDF]

				M. F. Ibrahim, D. Paparella, J. Ivanov, M. R. Buchanan, and S. J. Brister Gender-related differences in morbidity and mortality during combined valve and coronary surgery J. Thorac. Cardiovasc. Surg., October 1, 2003; 126(4): 959 - 964. [Abstract] [Full Text] [PDF]

				N. D. Desai and G. T. Christakis Stented Mechanical/Bioprosthetic Aortic Valve Replacement Card. Surg. Adult, January 1, 2003; 2(2003): 825 - 856. [Full Text]

				G. Rizzoli, L. Schiavon, and P. Bellini Does the use of bilateral internal mammary artery (IMA) grafts provide incremental benefit relative to the use of a single IMA graft? A meta-analysis approach Eur J Cardiothorac Surg, November 1, 2002; 22(5): 781 - 786. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): James E. Lowe Walter G. Wolfe H. Newland Oldham, Jr Peter Van Trigt, III Donald D. Glower Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Gall, S. Articles by Glower, D. D. Search for Related Content PubMed PubMed Citation Articles by Gall, S., Jr Articles by Glower, D. D.