(Circulation. 1995;92:50-57.)
© 1995 American Heart Association, Inc.
Articles |
In-Hospital and Long-term Outcome After Reoperative Coronary Artery Bypass Graft Surgery
From the Divisions of Cardiology and Cardiothoracic Surgery, Emory University School of Medicine, Atlanta, Ga.
Correspondence to William S. Weintraub, MD, Division of Cardiology, Emory University Hospital, 1365 Clifton Rd NE, Atlanta, GA 30322.
 |
Abstract |
|---|
 Top Abstract IntroductionMethods Results Discussion References |
|---|
Background Increasingly over the past several years, patients have returned after coronary surgery for reoperative procedures, and the experience has become substantial. In this report, we describe immediate- and long-term outcomes after reoperative coronary artery bypass graft surgery.
Methods and Results The source of data was the clinical database at Emory University. The surgical procedure and statistical methods were standard. Data were collected prospectively and entered into a computerized database. Follow-up was by letter, telephone, or hospital records documenting additional events resulting in readmission. In-hospital correlates of survival were determined by logistic regression, and long-term correlates were determined by Cox model analysis. There were 2030 patients with a mean age of 61 and a mean of 7.8±4.1 years since the first surgery. The mean ejection fraction was close to 50%, and the majority had three-vessel or left main disease. Urgent or emergency surgery was required in 16.6%. The internal mammary was used in 60.1%. Q-wave myocardial infarctions occurred in just over 5%. Neurological events increased from 1.2% at less than age 50 to 4.1% at more than age 70. The hospital mortality increased from 5.7% at less than age 50 to 10% at more than age 70, with an overall rate of 7.0%. Mortality was 5.7% for elective, 10.9% for urgent, and 16.4% for emergency cases. Angina was noted at follow-up in 41.3%. Urgent or emergency surgery, reduced ejection fraction, hypertension, older age, and female sex were univariate and multivariate correlates of in-hospital death. Diabetes was a univariate correlate only. Five- and 10-year survival rates were 76% and 55%, respectively. Five- and 10-year myocardial infarction–free survival rates were 63% and 40%, respectively. By 12 years, few patients were free of cardiac events. The univariate and multivariate correlates of long-term mortality were older age, reduced ejection fraction, hypertension, diseased vessels, presence of diabetes, congestive failure, and emergency surgery, with a strong trend for female sex. The use of the internal mammary artery was not a correlate for long-term mortality.
Conclusions Patients undergoing reoperative procedures have higher mortality initially and at long term than patients undergoing a first procedure. Expected mortality based on covariates may help in the decision of whether to perform reoperative coronary artery bypass graft surgery.
Key Words: surgery • aging • bypass • mortality • morbidity
|
Introduction |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Coronary artery bypass graft surgery (CABG) may be limited by incomplete revascularization, graft failure, and progression of narrowing in the native coronary arteries. Each of these problems alone or, more likely, in combination may lead to the need for reoperative CABG, which was first described within a few years of the first coronary surgical procedures.1 The problems of progression and graft failure accelerate after approximately 8 years, leading to an increased incidence of reoperative procedures.2 Furthermore, reoperative CABG has been shown to be an increasing part of current surgical practice.3 4 In the present study, we examine the in-hospital and long-term outcomes of reoperative CABG.
|
Methods |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Patient Population
The study population comprised patients undergoing a first reoperation at Emory University Hospitals between 1975 and 1993. Patients with valve disease, congenital heart disease, or primary myocardial diseases and patients requiring aneurysmectomy were excluded. The total population was 2030 (mean age, 61±9 years), and 2011 (99%) had follow-up (mean, 4.3±3.2 years).
Surgical Technique
Extracorporeal circulation was instituted
by standard
techniques,5 and perfusion was maintained at 2.0 to 2.4
L · min-1 · m-2. Systemic
hypothermia
(30°C to 25°C), topical hypothermia, and cold potassium
cardioplegia were used for myocardial protection. Cardioplegic solution
was reinfused at 20- to 30-minute intervals to maintain an
intramyocardial temperature of <20°C. After the patient was weaned
from cardiopulmonary bypass, the chest was closed with
standard techniques. The patients were then transferred to the surgical
intensive care unit.
Definitions
Angina was defined by the Canadian
Cardiovascular Society classification.6
The historical variables of previous myocardial infarction,
diabetes mellitus, systemic hypertension, and prior myocardial
infarction were obtained for the patient. An artery was considered
stenotic if there was 
50% diameter narrowing of a main
coronary artery or any of its major branches. The number of
arteries narrowed was determined by a set algorithm.7
Urgent surgical status was defined as surgery necessary within 1 day
and emergency surgical status as surgery necessary within 1 hour
because of symptoms or hemodynamic instability. The
number of grafts was determined by the number of distal anastomoses. A
postoperative myocardial infarction was determined by the development
of new Q waves. A neurological event was a persistent change in
neurological function, including disorientation persisting for several
days.
Data Analysis
All data were prospectively collected and
entered into a
computerized database. Data are displayed as a percentage or mean±SD
where appropriate. Categorical variables were compared by
2, and continuous variables were
analyzed by ANOVA. Where data are displayed by decade,
P values refer to the total trend in the population by
decade. Long-term survival was determined for the entire population
(in-hospital deaths and survivors) by the Kaplan-Meier
method,8 and correlates of survival were determined by the
Cox proportional hazards method.9 Statistical testing was
performed with BMDP.
|
Results |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Clinical characteristics, by decade of age, are given in Table 1
. The percentage of women rose from 13.9% under age 50
to 20.7% over age 70. The majority had had a previous myocardial
infarction. Hypertension was present in approximately half the
patients, whereas three of four had class III or IV angina. Congestive
heart failure and diabetes were present in a minority of patients.
The time from the first CABG increased from 5.7 years under age 50 to
9.3 years over age 70. Angiographic characteristics are shown in Table
2. The mean ejection fraction was close to 50%, and a
sizable minority had ejection fractions of less 50%. The majority had
three-vessel or left main disease, with a small increase in the
severity of disease in older patients. Procedural characteristics and
outcome are displayed in Table 3. Urgent or emergency
surgery was required in a minority. The number of grafts placed
increased with age. The internal mammary was used in the majority.
Q-wave myocardial infarctions occurred in just over 5%. Neurological
events increased from 1.2% under age 50 to 4.1% over age 70. The
hospital mortality increased from 5.7% under age 50 to 10% over age
70. Angina was common at follow-up but decreased from 50.8% under
age 50 to 25.7% over age 70. There were 95 patients in this series
from before 1980, 376 from 1980 through 1984, 778 from 1985 through
1989, and 781 since 1990. Over these time periods, age increased from
53±8 to 64±9 years, ejection fraction fell from 57±13% to
48±13%,
the incidence of three-vessel or left main disease rose from 45.7%
to 77.1%, and the incidence of hypertension increased from 35.9% to
51.8% (all P<.001). There was no consistent
pattern to frequency of emergency surgery and no change noted in sex.
Despite relatively profound changes in patient population, there were
no differences noted in in-hospital outcomes of Q-wave myocardial
infarction, neurological events, or death over this period.
|
|
|
The clinical, angiographic, and procedural characteristics in Tables 1
through 3
were used to determine
the in-hospital mortality. The
univariate and multivariate odds ratios as
well as P values for in-hospital mortality are displayed in
Table 4. For ejection fraction and age, the odds ratios
refer to an increase of 1 unit, for example, age 62 to 63 years or
ejection fraction from 46% to 47%. Urgent and emergency surgery,
reduced ejection fraction, hypertension, older age, and female sex were
univariate and multivariate correlates. The
univariate and multivariate odds ratios
were similar. Diabetes was a univariate correlate only. For
elective surgery, the mortality was 5.7%; for urgent surgery, 10.9%;
and for emergency surgery, 16.4%.
|
Long-term survival; myocardial infarction–free survival;
freedom from death, myocardial infarction, and a third coronary
surgery; and survival free of myocardial infarction, a third
coronary surgery, and angioplasty are shown as a layered set of
survival curves in Fig 1. There is an initial dip in the
curves representing in-hospital events; thereafter, the
curves become quite linear. Although survival may seem somewhat
reassuring at 76% at 5 years, it fell to 55% at 10 years. Myocardial
infarc-tion was the other major event. The additional influence of
further revascularization procedures seemed
somewhat less. Nevertheless, by 12 years, few patients were free of
cardiac events. Fig 2 shows the freedom from additional
revascularization procedures, which appear more
prominent on this figure than on Fig 1 because patients may
have had
other events, thereby masking the incidence of additional procedures.
Freedom from a third coronary
revascularization procedure, coronary
angioplasty, and additional revascularization of
either type are shown. Note that additional procedures are at first
unusual, but that after 6 to 8 years, repeat
revascularization becomes more common.
|
|
The clinical, angiographic, and procedural characteristics in Tables 1
through 3 were used to determine
long-term correlates of mortality.
The univariate and multivariate hazard
ratios as well as P values for long-term mortality are
given in Table 5. For ejection fraction and age, the
hazard ratios are for an increase of 1 unit. The univariate
and multivariate correlates were older age, reduced
ejection fraction, hypertension, diseased vessels, presence of
diabetes, congestive heart failure, and urgent or emergency surgery
with a strong trend for female sex. The use of the internal mammary
artery was not a correlate of long-term mortality. The hazard ratio
of 1.04 per year of age may also be expressed as 1.49 for each decade
increase in age (95% confidence interval, 1.32 to 1.69). Similarly,
patient characteristics were used to determine correlates of additional
procedures. The only multivariate correlates of
additional procedures were male sex (P=.0072) and younger
age (P=.0076). Men had a hazard ratio of 1.67 (95%
confidence interval, 1.12 to 2.49) for additional procedures, and the
hazard ratio was 1.21 (95% confidence interval, 1.05 to 1.39) for each
decade decrease in age.
|
Survival divided by age groups is shown in Fig 3. There
was little difference between patients under age 50 and between ages 50
and 59. Thereafter, mortality rose significantly. Although few patients
over age 70 were alive after 10 years, survival longer than this for
patients over age 70 undergoing reoperative surgery should not be
expected. Note that patients over age 70 were 9.3 years from their
original surgery and thus by 12 years would be over age 80 and more
than 20 years from their original surgery. Survival, divided by
ejection fraction 50%, 35% to 49%, and <35%, is shown in Fig
4. Note that the difference between the curves is
largely established in the first 1 to 2 years, and thereafter the
curves continue to separate more slowly. Survival divided by the
presence or absence of hypertension is shown in Fig 5.
Here, the curves are initially close together but splay out and become
more separated over time. Survival divided by vessels diseased is
displayed in Fig 6. Although the curves do not separate
as cleanly as in several of the other figures, a progression of
increased mortality from single to left main disease may be noted.
Survival divided by the absence and presence of diabetes is
presented in Fig 7. Similar to the curves for
hypertension, the curves for diabetes splay out over time, so that
survival by 10 to 12 years for patients with diabetes is very poor.
Survival divided by the presence or absence of congestive heart failure
is shown in Fig 8. These curves separate rapidly but
then fall in parallel, similar to those for ejection fraction but with
less separation. Survival divided by elective, urgent, or emergency
surgery is shown in Fig 9. The separation is essentially
immediate, reflecting initial in-hospital mortality.
Thereafter, the curves begin to come together, so that by 7 or 8 years
the curves overlap. Survival by sex is displayed in Fig 10.
Women have somewhat lower survival rates, although
the curves are close.
|
|
|
|
|
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
In the present study, the immediate- and long-term outcomes after reoperative coronary surgery have been presented in a large patient population. The in-hospital mortality was 7%, and 1-, 5-, and 10-year mortality was 11%, 23%, and 45%. The major correlate of in-hospital mortality was emergency surgery, and of long-term mortality, older age. The other major complications in-hospital were acute Q-wave myocardial infarction in 5.6% and neurological events in 2.8%. The constancy of in-hospital results despite an older and more severely diseased population in recent years suggests gradually improving techniques. The high in-hospital mortality with emergency surgery suggests that control or prevention of perioperative ischemia may be useful in lowering in-hospital mortality. There was a continuing incidence of myocardial infarction after hospital discharge as well as additional revascularization procedures.
The in-hospital mortality rate was higher and long-term survival rate was lower than in multiple series of first revascularization procedures. In a recent publication examining long-term outcome from Emory University, the in-hospital mortality was 1.0%, and 5- and 10-year survival rates were 91% and 78%, respectively. Five- and 10-year myocardial infarction–free survival rates were 83% and 65%, respectively. The patients in that series were younger and had less severe angina, fewer previous myocardial infarctions, less hypertension, less congestive heart failure, less diabetes, less severe coronary disease, and higher ejection fractions than those in the present study. In the randomized portion of CASS, the 5- and 10-year survival rates were 95% and 82% in the surgical arm,10 respectively. The patients in CASS were younger and had less angina and less severe disease. Van Brussel et al11 noted 5- and 10-year survival rates of 94% and 82%, respectively. These patients were also younger and had less severe disease, less diabetes, less hypertension, and less heart failure than those in the present study. Rahimtoola et al12 reported 5- and 10-year survival rates of 88% and 73%, respectively. These patients were also younger and had less diabetes, less hypertension, and fewer previous myocardial infarctions than the patients in this study. Loop et al13 also noted lower long-term mortality rates for first-time surgery and for patients who were younger and more often male and had less severe angina and less severe anatomic disease. Lowrie et al14 noted 5-year survival rates of 80% to 85% and 10-year survival rates of 65% to 70%. The patients were younger and had less severe disease, better left ventricular function, and less diabetes than in this series. Similar results were also noted in the European Cooperative study15 and the VA Cooperative study16 and in results from the Duke University database.17
Compared with a wealth of data concerning first-time coronary surgery, the data on reoperative surgery are more scant. Kirklin and Barratt-Boyes18 noted that in-hospital mortality after coronary surgery is approximately twice that of first surgery. In the CASS registry experience, 9086 patients having a first procedure were compared with 283 patients having a reoperative procedure.19 The mortality and myocardial infarction rates after a first procedure were 3.1% and 6.4% versus 5.3% and 5.8%, respectively, for a reoperative procedure. The patients undergoing reoperation were younger (mean age, 52±9 years in the CASS experience) and had less severe anatomic disease than in the present report. In the largest series, Loop et al20 reported on results in 2509 patients. Mortality ranged from 2% to 5%, and new Q-wave myocardial infarctions ranged from 4% to 8%, with improved results toward the end of the series. For hospital survivors, the 10-year survival rate was 69.3%, and event-free survival was 41.2%. Thus, these patients probably have not had as good a prognosis as first-time surgical patients. In addition, these patients were younger and had less diabetes and less hypertension than those in the present series. In contrast to the present study, Loop et al13 noted improved survival in patients with internal mammary artery grafts, although this was not as important as in their study of first surgeries.
There is an alternative to coronary surgery for at least some patients in the form of interventional procedures in the catheterization laboratory. In a recent series from Emory University of vein graft angioplasty, the in-hospital mortality was 1.2%, Q-wave myocardial infarction rate was 2.2%, and rate of need for emergency coronary surgery was 3.5%.21 Restenosis in vein grafts is probably a more severe problem than in native vessels, and new devices such as atherectomy and stents may lower the rate of restenosis. There are little comparative data on the devices, and the little that there are show little difference in outcome.22 There are little data comparing interventional procedures with reoperative surgery. In a recent preliminary, observational comparison, the long-term results of interventional procedures were superior to those of reoperative surgery, but the selection bias could not be accounted for.23 There are no randomized data and no trial in planning as far as we know. It is not at all certain that the patients treated with interventional procedures are similar in clinical or angiographic characteristics to patients undergoing reoperative surgery. Although the patients undergoing surgery would be expected to be older and to have more severe disease, this may not always be the case, as some patients with very severe disease may be referred to angioplasty as a last-ditch rescue.
How, then, are patients to be cared for after coronary surgery? There certainly are data on increasing incidence of reoperative procedures after a period of years, suggesting that progression of native vessel disease, graft failure, or both accelerate after approximately 8 years.2 24 25 In patients with recurrent symptoms unresponsive to medical therapy, referral for revascularization is reasonable. Deciding how to follow and when to catheterize and then revascularize asymptomatic patients with positive noninvasive testing and angiographic evidence of ischemic potential is much more difficult. The decision must be tempered with the knowledge that these patients have severe disease and will continue to have problems after palliative surgery. Multivariate models such as those developed in the present study may be used to more accurately determine both immediate- and long-term prognosis. Nevertheless, there is no adequate substitute for good comparative studies that will permit the assessment of which patients are needing additional revascularization and of the overlapping patients who are suitable for either catheter-based or surgical procedures. Finally, studies are needed comparing the costs in relation to the clinical outcomes of the various forms of revascularization after coronary surgery.
|
References |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
1. Adam M, Geisler GF, Lambert CJ, Mitchel BF Jr. Reoperation following clinical failure of aorta-to-coronary artery bypass vein-grafts. Ann Thorac Surg. 1972;14:272-281. [Medline] [Order article via Infotrieve]
2. Weintraub WS, Jones EL, Craver JM, Guyton RA. Frequency of repeat coronary bypass or coronary angioplasty after coronary artery bypass surgery using saphenous venous grafts. Am J Cardiol. 1994;73:103-112. [Medline] [Order article via Infotrieve]
3. Jones EL, Weintraub WS, Craver JM, Guyton RA, Cohen CL. Coronary bypass surgery: is the operation different today? J Thorac Cardiovasc Surg. 1991;101:108-115. [Abstract]
4. Christakis GT, Ivanov J, Weisel RD, Birnbaum PL, David TE, Salerno TA, and the Cardiovascular Surgeons of the University of Toronto. The changing pattern of coronary artery bypass surgery. Circulation. 1988;80(suppl I):I-151-I-161.
5. Jones EL, Craver JM, King SB III, Douglas JS, Brown CM, Bone DK, Hatcher CR Jr. Clinical, anatomic and functional descriptors influencing morbidity, survival and adequacy of revascularization following coronary bypass. Ann Surg. 1980;192:390-402. [Medline] [Order article via Infotrieve]
6. Campeau L. Grading of angina pectoris. Circulation. 1975;54:522-523. Letter.
7.
Talley JD, Hurst JW, King SB III, Douglas JS, Roubin
GS, Gruentzig AR, Anderson HV, Weintraub WS. Clinical outcome 5
years after attempted percutaneous transluminal
coronary angioplasty in 427 patients.
Circulation. 1988;77:820-829.
8. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc. 1958;53:457-481.
9. Cox DR. Regression models and life tables. J R Stat Soc Ser B.. 1972;34:187-202.
10.
Alderman EL, Bourassa MG, Cohen LS, Davis KB, Kaiser
GC, Killip T, Mock MB, Pettinger M, Robertson TL, for the CASS
Investigators. Ten-year follow-up of survival and
myocardial infarction in the randomized coronary artery surgery
study. Circulation. 1990;82:1629-1646.
11. van Brussel BL, Plokker HWT, Ernst SMPG, Ernst NM, Knaepen PJJ, Koomen EM, Tijssen JGP, Vermeulen FEE, Voors AA. Circulation. 1993;88(pt 2):87-92.
12. Rahimtoola SH, Fessler CL, Grunkemeier GL, Starr A. Survival 15 to 20 years after coronary bypass surgery for angina. J Am Coll Cardiol. 1993;21:151-157. [Abstract]
13. Loop FD, Lytle BW, Cosgrove DM, Stewart RW, Goormastic M, Williams GW, Golding DM, 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. [Abstract]
14. Lowrie GM, Morris GC Jr, Earle N. Long term results of coronary bypass surgery. Ann Surg.. 1991;213:377-385. [Medline] [Order article via Infotrieve]
15. Varnauskas E, and the European Coronary Surgery Study Group. Twelve-year follow-up of survival in the randomized European coronary surgery study. N Engl J Med. 1988;319:332-337. [Abstract]
16. The Veterans Administration Coronary Artery Bypass Surgery Cooperative Study Group. Eleven-year survival in the Veterans Administration randomized trial of coronary bypass surgery for stable angina. N Engl J Med.. 1984;311:1333-1339. [Abstract]
17. Smith LR, Harrell FE, Rankin JS, Califf RM, Pryor DB, Muhlbaier LH, Lee KL, Mark DB, Jones RH, Oldham HN, Glower DD, Rerves JG, Sabiston DC Jr. Determinants of early versus late cardiac death in patients undergoing coronary artery bypass graft surgery. Circulation. 1991;84(suppl III):III-245-III-253.
18. Stenotic arteriosclerotic coronary artery disease. In: Kirklin JW, Barratt-Boyes BG, eds. Cardiac Surgery. New York, NY: Churchill-Livingstone; 1993:333.
19. Foster ED, Fisher LD, Kaiser GC, Myers WO, Principal Investigators of CASS and Their Associates. Comparison of operative mortality and morbidity for initial and repeat coronary artery bypass grafting: the Coronary Artery Surgery Study (CASS) registry experience. Ann Thorac Surg.. 1984;38:563-570.[Abstract]
20. Loop FD, Lytle BW, Cosgrove DM, Woods EL, Stewart RW, Golding LAR, Goormastic M, Taylor PC. Reoperation for coronary atherosclerosis: changing practice in 2509 consecutive patients. Ann Surg.. 1990;212:378-384. [Medline] [Order article via Infotrieve]
21. Douglas JS, King SB III. Ten-year follow-up of patients undergoing vein graft angioplasty. Circulation. 1994;90(suppl I):I-333. Abstract.
22. Waksman R, Thomas C, Ghazzal ZMB, Weintraub WS, Shen Y, Douglas JS Jr. Palmaz-Schatz stent (PSS), directional coronary atherectomy (DCA), and balloon angioplasty for saphenous vein graft stenoses (SVG). Circulation. 1994;90(suppl I):I-334. Abstract.
23. Dahiya RS, O'Keefe JH, Ligon R, McCallister BD. PTCA vs reoperation for patients with prior bypass surgery. Circulation. 1994;90(suppl I):I-334. Abstract.
24. Cosgrove DM, Loop FD, Lytle BW, Gill CC, Golding LAR, Gibson C, Stewart RW, Taylor PC, Goormastic M. Predictors of reoperation after myocardial revascularization. J Thorac Cardiovasc Surg.. 1986;92:811-821.[Abstract]
25. Bourassa MG, Enjalbert M, Campeau L, Lesperance J. Progression of atherosclerosis in coronary arteries and bypass grafts: ten years later. Am J Cardiol. 1984;53:102C-107C.[Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  P. Sergeant, E. Blackstone, B. Meyns, B. Stockman, and R. Jashari First cardiological or cardiosurgical reintervention for ischemic heart disease after primary coronary artery bypass grafting Eur. J. Cardiothorac. Surg., November 1, 1999; 14(5): 480 - 487. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Craver, G. T. Hodakowski, Y. Shen, W. S. Weintraub, K. D. Accola, R. A. Guyton, and E. L. Jones Third-Time Coronary Artery Bypass Operations: Surgical Strategy and Results Ann. Thorac. Surg., December 1, 1996; 62(6): 1801 - 1807. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||