- Email: [email protected]
Combined myocardial revascularization and carotid endarterectomy
J THoRAc
CARDIOVASC SURG
85:577-589, 1983
Combined myocardial revascularization and carotid endarterectomy Operative and late results in 331 patients Simultaneous coronary artery bypass and carotid endarterectomy were performed in 331 patients (mean age 61 years) at the Cleveland Clinic from 1973 through 1981, Of these, 195 (59%) had Functional Class lll-IV angina pectoris, 308 (93%) had multiple-vessel coronary artery disease (CAD), 68 (21%) had over 50% stenosis of the left main coronary artery, and 185 (56%) had either segmental or diffuse impairment of left ventricular function, Asymptomatic carotid stenosis was documented in 173 patients (52'7c), and the remaining 158 had experienced either previous transient cerebral ischemia (38%) or completed strokes (10%). Single aorta-coronary grafts were placed in 59 patients (18%), double gratis in 131 (40%), and three or more grafts in 141 (42%). Nineteen patients (5.7%) died postoperatively in the hospital, Neurologic deficits occurred in 30 patients (9.0%) and produced permanent functional impairment in 15 (4.5%). Late results have been obtained for 312 operative survivors at a mean postoperative interval of 38 months. Thirty-eight patients (12%) have died, but the 5 year life-table survival rate of the study group was identical to that of the normal population aged 61 years, Significant differences in cumulative 5 year survival rates were identified among diabetic patients (p < 0.025) and among those receiving single rather than double (p < .005) or multiple (p < .01) coronary grafts. Although 18 patients (5.8%) have had late strokes, only five (1.6%) of these strokes have involved the cerebral hemisphere on the same side as combined carotid endarterectomy,
Norman R, Hertzer, M,D., Floyd D. Loop, M.D., Paul C. Taylor, M.D., and Edwin G. Beven, M.D., Cleveland, Ohio
Myocardial infarction and stroke are the two leading causes of cardiovascular death in the United States. It is not surprising, therefore, that severe, simultaneous coronary and carotid artery disease may be documented in some patients initially under consideration for either elective myocardial revascularization or extracranial arterial reconstruction alone. Although multifocal atherosclerosis is common to all such patients, their referral .for surgical management usually assumes one of two distinctly different courses. First, a relatively small subset of those with known coronary artery disease (CAD) is found to have adFrom the Cleveland Clinic Foundation, The Department of Thoracic and Cardiovascular Surgery and The Department of Vascular Surgery, Cleveland, Ohio. Received for publication May 24, 1982. Accepted for publication July 2, 1982. Address for reprints: Norman R. Hertzer, M.D., Department of Vascular Surgery, The Cleveland Clinic Foundation, 9500 Euclid Ave., Cleveland, Ohio 44106.
vanced carotid artery lesions preceding elective myocardial revascularization. The incidence of postoperative stroke after aorta-coronary bypass using contemporary methods for extracorporeal perfusion usually does not exceed 2%,1-4 and many of these neurologic deficits are manifest as diffuse encephalopathy or intellectual impairment that does not specifically implicate extracranial occlusive disease. Nevertheless, hemispheric neurologic complications might be expected to occur with greater frequency among patients with recognized carotid stenosis, and several previous reports have recommended that carotid endarterectomy should be performed in selected patients as a staged or combined procedure at the time of myocardial revascularization :;-11 Second, coronary atherosclerosis is prevalent among patients who have carotid disease that is sufficiently severe to warrant surgical correction. Provided the operative stroke rate is not excessive, myocardial infarctions are principally responsible for the few postoperative deaths that occur after carotid endarterectomy.P: 13 577
The Journal of Thoracic and Cardiovascular
578 Hertzer et al.
Surgery
Table I. Extent and distribution of coronary artery disease in 331 patients Coronary artery disease >50%sterwsis Distribution
No.
Left coronary artery Main trunk Anterior descending Circumflex Right coronary artery Single vessel Double vessel Triple vessel
68 286 236 273 23 98 210
I
> 70% stenosis
%
No.
21 86 71 82 7 30 63
39 244 191 245 51 131 135
I
%
12 74 58 74 16 40 40
Moreover, myocardial infarctions account for as many as 70% of all late deaths following successful extracranial arterial reconstruction and are especially common among patients suspected to have CAD on clinical grounds. 13-19 The incidence of fatal myocardial infarction during the late postoperative interval after carotid endarterectomy at the Cleveland Clinic has been shown to equal or surpass the total incidence of both fatal and nonfatal stroke in all age groups.s? In an attempt to identify those patients with cerebrovascular disease who have severe, associated CAD, coronary angiography has been routinely employed before elective carotid endarterectomy at this institution since 1978. 21 In a previous report, we described early postoperative results for the surgical management of simultaneous carotid and coronary vascular disease utilizing staged operations in 59 patients and a combined approach in 115 others. 6 The present investigation is an analysis of intermediate-term results as well as postoperative complications for a series of 331 patients who required combined myocardial revascularization and carotid endarterectomy at the Cleveland Clinic from 1973 through 1981.
Patients and metbods The 331 patients in this series represent only 1.7% of 22,100 patients who had direct myocardial revascularization at the Cleveland Clinic during the period of this study. There were 259 men (78%) and 72 women (22%), ranging in age from 39 to 85 years (mean 61 years). Nineteen (6%) were less than 50 years of age, 108 (32%) were 50 to 59 years of age, 145 (44%) were 60 to 69 years of age, and 59 (18%) were over 70 years of age. In comparison, the mean age of all 22,100 patients undergoing aorta-coronary bypass from 1973 through 1981 was 54 years.
Atherosclerotic risk factors. Diabetes mellitus, defined as known diabetes under medical treatment or disclosed by routine preoperative studies (abnormal glucose tolerance test, fasting blood sugar > 140 mg/dl) was present in 79 patients (24%). One hundred sixty-three patients (49%) had hypertension under medical management or had sustained blood pressure measurements above 150/90 mm Hg. One hundred seventyeight patients (54%) either smoked cigarettes or had discontinued their chronic use of tobacco within the previous 5 years. Serum cholesterol levels ranged from 95 to 541 mg/dl (mean 267 mg/dl), and serum triglyceride levels ranged from 51 to I ,223mg/dl (mean 216 mg/dl).· Patient selection. Patients scheduled for myocardial revascularization during this study period were questioned specifically concerning possible symptoms of previous completed strokes or transient cerebral ischemic attacks (TIA). Auscultation for carotid bruits was a routine feature of physical examination. Carotid angiography was performed for patients with previous neurologic symptoms and for those with asymptomatic cervical bruits that appeared to originate at the carotid bifurcation, particularly when the results of noninvasive assessment of internal carotid blood flow were abnormal.P Documentation of carotid disease was obtained by the transfemoral (Seldinger) approach, by cineangiography at the time of cardiac catheterization, or by digital subtraction angiography for all patients in this series. Carotid endarterectomy was recommended for patients who had previous neurologic symptoms and appropriate carotid bifurcation disease, as well as for those having asymptomatic bruits caused by stenosis of the cervical segment of the internal carotid artery that reduced luminal diameter by 70% or more. Asymptomatic stenosis compromising the residual lumen by less than 70% was occasionally repaired in the presence of contralateral internal carotid occlusion. Preoperative cardiac status. Symptoms. As defined by the New York Heart Association, Functional Class II angina pectoris was present in 106 patients (32%), Class III in 132 (40%), and Class IV in 63 (19%). Thirty patients (9%) had no angina pectoris at the time of their initial investigations, but 32 others (10%) had experienced symptoms of congestive heart failure. Electrocardiography. A standard, twelve-lead electrocardiogram (ECG) was normal in 92 patients (28%). Eighty-three patients (25%) had Q waves confirming previous myocardial infarctions, and III others (33%) had ischemic ST-T segment changes. Five patients (2%) had other ECG abnormalities such as left bundle
Volume 85 Number 4 April, 1983
Coronary bypass and carotid endarterectomy
100 65
579
PATIENTS O ALLIN=22,100)
60 55
•
COMBINED GROUP (N=331)
50 45
....
...z u
......'"
40 35 30 25 20 15 10 5
CORONARY ARTERY DISEASE
VENTRICULAR FUNCTION
PERIOPERATIVE COMPLICATIONS
Fig. 1. Graphic representation of the prevalence of preoperative risk factors and the incidence of postoperative complications for the 331 patients in this report and for the total of 22,100 patients undergoing myocardial revascularization during this period of study.
branch block, and 40 (12%) had two or more ECG findings. Cardiac catheterization. Table I presents a summary of data obtained during coronary angiography for all patients in this report, according to whether the maximum degree of angiographic stenosis exceeded either 50% or 70% of luminal diameter. The left main coronary artery was more than 70% stenotic in 39 patients (12%) and the anterior descending branch in 244 (74%). The presence of high-grade, subtotal coronary artery stenosis dictated a combined operation rather than a staged approach to associated carotid disease in many of these patients. More than 50% stenosis was found in a single coronary artery in 23 patients (7%), two coronary arteries in 98 (30%), and three coronary arteries in 210 (63%). According to the same criterion of 50% luminal stenosis, 14% of the total of 22,100 patients undergoing direct myocardial revascularization during this study period had single-vessel coronary disease, 30% had double-vessel disease, and 56% had triple-vessel disease. Cine left ventriculography demonstrated normal function in 146 patients (44%). Segmental akinesia was found in 155 patients (46%), and 30 (9%) had diffuse left ventricular impairment. In comparison, 52% of all 22,100 patients had normal ventriculograms, and 48% had some element of ventricular dysfunction.
Graphic representation of the extent of CAD (> 50% luminal stenosis) and left ventricular impairment for the series of 331 patients in this report and for the group of 22,100 patients from whom they were selected is presented in Fig. 1. The greater prevalence of triple-vessel disease (p < 0.0l), of left main coronary involvement (p < 0.00l), and of ventricular impairment (p < 0.01) in patients requiring combined procedures was statistically significant.
Preoperative carotid status. Symptoms. A history of previous neurologic symptoms was obtained from 158 patients (48%). Of these, 124 (38%) had experienced prior hemispheric or vertebrobasilar TIA, and 34 (10%) had had completed strokes. The remaining 173 patients (52%) had cervical bruits and were found to have severe, asymptomatic carotid stenosis during cerebral angiography. Angiographic findings. A summary of the extent and distribution of carotid artery disease determined by angiography is given in Table II. Only five patients (2%) had less than 50% stenosis of the internal carotid artery that was repaired during combined procedures, and each had ulcerated atheromatous disease as well as previous neurologic symptoms that could have been caused by microemboli arising from such lesions. The contralateral internal carotid artery was less than 50% stenotic in 216 patients (65%). Fifteen of these patients
The Journal of
580 Hertzer et al,
Thoracic and Cardiovascular Surgery
Table II. Extent and distribution of carotid artery disease in 331 patients Distribution Ipsilateral Carotid artery disease Stenosis <50% 50%·90% >90% Occlusion
No.
5 245 81
0
I
Contralateral
%
No.
2 74 24
216 69 9
37
I
%
65 21 3 II
had undergone contralateral carotid endarterectomy as a remote or staged procedure. Forty-six other patients (14%), however, had over 90% stenosis or occlusion of the contralateral internal carotid artery. Surgical management. Early in the study period, median sternotomy customarily was performed as the initial procedure so that prompt cannulation for cardiopulmonary bypass could be done if myocardial function deteriorated during carotid endarterectomy. Because of advances in anesthetic techniques and the pharmacologic support of cardiac performance, carotid endarterectomy in recent years has been completed before sternotomy with substantial savings in operative blood loss. Since harvesting and preparation of the saphenous vein for aorta-coronary grafting have usually been done at the same time as carotid endarterectomy, combined procedures have extended operating time only 45 minutes to I hour beyond that required for myocardial revascularization alone. Complete hemostasis and closure of the cervical incision was always delayed until intraoperative anticoagulation had been reversed and sternotomy closure had been completed. Coronary artery procedures. The technical details of direct myocardial revascularization at the Cleveland Clinic during the period of this study have been presented elsewhere.P: 23 Cold potassium cardioplegia was introduced in 1978 and, except for some patients with single-vessel disease, has been used in almost all revascularization operations since that time. Fifty-nine (18%) of the 331 patients in the present series had single aorta-coronary grafts, 131 (40%) had double grafts, 113 (34%) had triple grafts, and 28 (8%) had four or more grafts. Complete revascularization of all significant coronary artery lesions was accomplished in 197 patients (60%). In comparison, 28% of the total of 22,100 patients undergoing direct myocardial revascularization during this study period had single grafts, 38% had double grafts, 25% had triple grafts, and 9% had four or more grafts. In addition to carotid endarterectomy, 16 patients in
this report (4.8%) required other combined procedures, including aortic or mitral valve replacement in 12, open mitral commissurotomy in one, ventricular aneurysmectomy in two, and resection of a symptomatic abdominal aortic aneurysm in one. Carotid artery procedure. Technical aspects of carotid endarterectomy at this institution also have previously been described.v' Every patient received systemic sodium heparin, 5,000 to 10,000 international units, prior to the application of carotid clamps. An indwelling carotid shunt was inserted in 274 patients (83%), and only three of 185 operations since 1978 have been performed without shunts. Perfusion of the ipsilateral internal carotid artery was rarely interrupted for longer than 3 to 4 minutes during any operation done with the use of a shunt. Tacking sutures were frequently employed to secure the distal intima of the internal carotid artery, but arteriotomy closure required venous patch angioplasty in fewer than 5% of patients. Data analysis. The Fisher exact test of the null hypothesis was employed for analysis of paired variables. Differences in life-table data were evaluated by the method described by Mantel."
Early results Postoperative mortality. Nineteen of the 331 patients died in the hospital during the postoperative period, for an operative mortality rate of 5.7%. Although five patients eventually had multisystem failure, the principal cause of postoperative death was acute myocardial infarction or left ventricular failure in nine patients (2.7%), acute stroke or respiratory insufficiency in three each (0.9%), sepsis in two (0.6%), and renal failure or ruptured abdominal aortic aneurysm in one each (0.3%). Age and sex. No patient less than 50 years of age died postoperatively. Fatal complications occurred in eight (7.4%) of those 50 to 58 years of age, in five (3.4%) of those 60 to 69 years of age, and in six (10%) of those over 70 years of age. Mortality rates were 6.3% for 127 patients less than 60 years of age and 5.4% for 204 patients over 60 years of age. This difference was not statistically significant. Although postoperative death was more common among men (6.9%) than among women (1.4%), this trend was not firmly supported by statistical analysis (0.05 < p < 0.1). Diabetes and hypertension. No meaningful differences in hospital mortality rates were identified between diabetic (7.6%) and nondiabetic patients (5.2%) or between patients with hypertension (6.7%) and those who were normotensive (4.8%). Preoperative cardiac status. Neither the severity of
Volume 85 Number 4
Coronary bypass and carotid endarterectomy
581
April,1983
preoperative cardiac symptoms nor the extent and distribution of CAD had a significant influence upon postoperative mortality. Although mortality rates for patients with Functional Class I-II and Class III-IV angina pectoris were 2.9% and 7.7%, respectively, this difference was not statistically valid (0.05 < P < 0.1). Likewise, there were no perceptible differences in postoperative mortality among patients having singlevessel, double-vessel, or triple-vessel CAD (p > 0.1). However, postoperative mortality was higher among patients with segmental (7. 1%) or diffuse (13%) ventricular impairment than in those having normal preoperative ventricular function (2.7%). Four postoperative deaths occurred among 146 patients (2.7%) who had normal left ventriculograms, whereas there were 15 deaths in a total of 185 patients (8. 1%) with ventricular dysfunction (p < 0.05). Surgical management. There was only one postoperative death (1.7%) in the subset of 59 patients who received a single aorta-coronary bypass graft, and this death was caused by an acute myocardial infarction in a patient with additional severe but surgically uncorrectable CAD. Fatal complications occurred in seven (5.3%) of 131 patients who had double grafts, in nine (8.0%) of 113 who had triple grafts, and in two (7.1 %) of 28 who had four or more grafts. Twelve (6.1 %) of 197 patients died after complete revascularization of all significant coronary lesions, and there were seven deaths in the subset of 134 patients (5.2%) who received only incomplete revascularization. None of these differences was statistically meaningful. Throughout the period of this study, 21 patients (6.3%) had perioperative myocardial infarctions documented by serum glutamic oxaloacetic transaminase levels greater than 100 international units or by new Q waves on the postoperative ECG. Eight (39%) of these infarctions were associated with a fatal outcome. Preoperative carotid status. Thirteen (8.2%) of 158 patients with previous neurologic symptoms and six (3.5%) of 173 patients with asymptomatic carotid stenosis died after combined operations. There were eight deaths among 115 patients (7.0%) who had more than 50% stenosis or occlusion of the contralateral internal carotid artery and 11 deaths among 216 patients (5.1%) with less than 50% contralateral carotid stenosis. None of these differences was statistically significant. Postoperative neurologic complications. Thirty (9.0%) of the 331 patients had some degree of perioperative neurologic deficit as a direct complication of their combined operations. Sixteen of these deficits (4.8%) were manifest by symptoms and signs appropriate to the ipsilateral cerebral hemisphere, 10 (3.0%) to the contralateral cerebral hemisphere, and four
Table III. Technical considerations and results during consecutive periods of study 1973-1977 (N = 146) Factors and results
No.
Coronary artery Single bypass graft Double grafts Three or more grafts Periop. myocardial infarction Operative mortality Carotid artery Intraop. shunting Postop. neurologic deficit
I
%
1978-1981 (N = 185) No.
I
%
35 67 44 13 8
24 46 30 8.9 5.5
24 64 97 8 II
13 35 52 4.3 5.9
90
62 12.3
182 12
98 6.6
18
Table IV. Principal causes of late death among 3 J2 operative survivors
Principal cause oflate death
No.
Percent of late deaths
Percent of operative survivors
5 5
40 10 8 5 5 3 3 13 13
4.8 1.3 1.0 0.6 0.6 0.3 0.3 1.6 1.6
38
100
12.1
Myocardial infarction Stroke Ruptured aortic aneurysm Congestive heart failure Malignant neoplasm Chronic renal disease Pulmonary embolism Other Unknown
15 4 3 2 2
Totals
I I
(1.2%) to the brain stem. A total of 15 deficits (4.5%) were temporary, including 13 that resolved within a few days postoperatively and two completed strokes requiring a maximum of 2 months for complete recovery. Fifteen other completed strokes (4.5%) produced some element of permanent functional disability. Seven of the permanent strokes were associated with only mild impairment. Eight others were severe, and three of these (0.9%) were fatal. None of the postoperative neurologic deficits occurred in any patient less than 50 years of age. There were nine deficits among those 50 to 59 years of age (8.3%), 13 among those 60 to 69 years of age (9.0%), and eight among those over 70 years of age (14%). Neurologic deficits occurred in 7. 1% of patients less than 60 years of age and in 10% of those over 60 years of age. None of these differences was statistically significant. Likewise, there was no statistically valid trend in the incidence of postoperative neurologic deficits between diabetic (15%) and nondiabetic patients (7.1 %), between hypertensive patients (9.2%) and those with-
The Journal of Thoracic and Cardiovascular Surgery
582 Hertzer et al.
Table V. Complete life-table survival data, including operative mortality
Postop. year
No. alive at beginning of the year
No. lost to follow-up during the year
No. alive observed for only part of the year
No. exposed to risk of dying during the year
No. dying during the year
Proportion dying during the year
Proportion surviving the year
Proportion surviving to end ofeach year
30 days I 2 3 4 5 6 7 8 9
331 312 245 198 153 100 82 41 18 5
0 0 0 0 0 0 0 0 0 0
0 57 41 42 45 14 36 21 13 5
331 283.5 224.5 177 130.5 93 64 30.5 11.5 2.5
19 10 6 3 8 4 5 2 0 0
0.057 0.035 0.027 0.017 0.061 0.043 0.078 0.066 0.000 0.000
0.943 0.965 0.973 0.983 0.939 0.957 0.922 0.934 1.000 1.000
0.943 0.910 0.885 0.870 0.817 0.782 0.721 0.673 0.673 0.673
out hypertension (8.9%), between patients having normal ventricles (7.5%) and those with impaired ventricular function (10%), or among those receiving one (4.3%), two (9.2%), or three or more (9.5%) aortacoronary bypass grafts. Indications for operation. Fourteen (8.1%) of 173 patients with asymptomatic carotid stenosis had postoperative neurologic deficits. Deficits occurred in 16 (10%) of 158 patients who had experienced previous neurologic symptoms, including nine deficits among those with prior TIA (7.3%) and seven among those with completed strokes (21%). Despite an apparent tendency for those with previous strokes to sustain a higher incidence of postoperative neurologic deficits, the number of patients with previous strokes was too small to support this trend by statistical analysis (0.05 < P < 0.1). Preoperative carotid status. No neurologic complications occurred in any of the five patients with previous neurologic symptoms who had less than 50% stenosis of the ipsilateral carotid artery. Twenty-five (10%) of the 245 patients with 50% to 90% stenosis of the ipsilateral internal carotid artery had postoperative neurologic deficits, as did five (6.2%) of 80 patients having over 90% ipsilateral stenosis. This difference was not significant. However, the status of the contralateral internal carotid artery imposed a meaningful difference in the incidence of postoperative neurologic complications. Deficits occurred in 6.9% of those with less than 50% stenosis of the contralateral internal carotid artery, in 8.7% of those with 50% to 90% stenosis, and in 20% of those with over 90% stenosis or occlusion of the contralateral internal carotid artery. The neurologic complication rate of patients with over 90% stenosis or internal carotid occlusion was significantly higher than that of patients with less than
50% stenosis of the contralateral internal carotid artery (p < 0.02). Carotid shunting. Early in the study period, 57 patients underwent combined carotid endarterectomy without the use of an intraoperative shunt. Postoperative neurologic deficits occurred in seven (12%) of these patients. Indwelling shunts have been employed in 274 procedures, and 23 (8.4%) of these patients experienced postoperative neurologic complications. This difference did not attain statistical significance. Trends in morbidity and mortality. The hospital mortality rate for all 22,100 patients who underwent myocardial revascularization from 1973 to 1981 was 1.9% (Fig. 1). Postoperative strokes occurred in 1.8%, and 2.8% experienced perioperative myocardial infarctions. The mortality rate for the 331 patients who had combined operations was 5.7%. Postoperative neurologic deficits occurred in 9.0%, of whom 4.5% had permanent deficits, and 6.3% had perioperative myocardial infarctions. All of these differences were highly significant (p < 0.001). Table III contains comparable data concerning surgical management and the incidence of major postoperative complications after combined procedures during two consecutive periods of study, 1973 to 1977 (146 patients) and 1978 to 1981 (185 patients). During these intervals, there were obvious trends favoring multiple aorta-coronary bypass grafts and the routine use of intraoperative carotid shunting. Together with advances in anesthetic methods, intraoperative myocardial protection, and the technology of postoperative care, these trends appear to be associated with substantial reduction in the incidence of perioperative myocardial infarction and neurologic deficit in the study group. This improvement has not yet achieved statistical significance, however, and the mortality rate for combined
Volume 85 Number 4 , April, 1983
Coronary bypass and carotid endarterectomy
583
100
90 80 70 ~
60
:c>
50
'"
40
s J '"
30 20
•
COMBINED CORONARY-CAROTID OPERATIVE SURVIVORS (N=312)
a
CAROTID ENDARTERECTOMY OPERATIVE SURVIVORS, 1969-1973 (N=142)
D. NORMAL MALE POPULATION (1976), AGE 61
10 2
3
4
5
POSTOPERATIVE YEARS
Fig. 2. Life-table survival curves for 312 patients in this report, for a normal male population of the same age,26 and for 142 patients with suspected but uncorrected coronary artery disease after carotid endarterectomy. 13
myocardial revascularization and carotid endarterectomy has remained relatively constant. Late results Late survival. Complete late information has been obtained for all 312 operative survivors at a mean postoperative interval of 38 months. Thirty-eight (12%) of these 312 patients have died at a mean follow-up interval of 33 months, including 26 (11%) of 241 men (mean 34 months) and 12 (17%) of71 women (mean 32 months). Table IV presents a summary of the principal causes of late death, insofar as dependable information was available from family members, referring physicians, or civil authorities. Cardiovascular diseases are known to have been responsible for 24 (63%) of the 38 deaths. Myocardial infarction has accounted for 40% of late deaths but may have been more common, since the cause of sudden death without other verification was considered as unknown. Complete life-table data, including operative mortality, for the 331 patients in this series are given in Table V. Annual late mortality has ranged from 1.7% to 7.8%, and the 5 year cumulative survival for 312 operative survivors presently is 83% ±6% (SD). Fig. 2 provides graphic representation of life-table survival data for the 312 operative survivors in this study, for a normal male population of the same age (61
years);" and for 142 patients (mean age 60 years) with suspected but undocumented CAD who survived carotid endarterectomy at the Cleveland Clinic from 1969 through 1973. 13 The late survival curve for patients who underwent combined myocardial revascularization and carotid endarterectomy in this report is identical to that of the normal population during the first 3 postoperative years. A slight difference in 5 year survival between these two groups is not statistically significant. Actuarial 5 year survival of patients with suspected CAD from the 1969 to 1973 series was 62% ± '13% (SD). The difference in 5 year survival between operative survivors in this group and those in the present series of combined operations was statistically valid (p < 0.0005). Diabetes and hypertension. The 312 patients in the follow-up group included 73 with diabetes. Life-table survival curves for these patients and for 239 others who did not have diabetes are shown in Fig. 3, A. Cumulative 5 year survival rates of 69% for diabetic and 87% for nondiabetic patients were statistically different (p < 0.025). At the time of their combined operations, 175 patients were defined as hypertensive, while 137 others did not have hypertension. Life-table survival rates for these two subsets were 84% and 81%, respectively (Fig. 3, B). This difference was not statistically significant.
The Journal of
584 Hertzer et al.
Thoracic and Cardiovascular Surgery
100 90 A.
0__
•
~--------o--o--
-.~.----o
80 70 60
~
• DIABETIC OPERATIVE SURVIVORS (N=73)
o NON-DIABETIC
•
OPERATIVE SURVIVORS (N=239)
~
C
0
'"
100
>
s
:::l
'" B.
90
4
5
-a~.
•
~~~--------.--
80 70 60 0
0-
0
4
5
• HYPERTENSIVE OPERATIVE SURVIVORS (N=17 S)
o NORMOTENSIVE OPERATIVE SURVIVORS (N=137)
2
3
POSTOPERATIVE YEARS
Fig. 3. Life-table survival curves for (A) diabetic patients and those without diabetes and (B) hypertensive patients and those without hypertension.
Ventricular function and myocardial revascularization, One hundred forty-two of the 312 patients in the follow-up group had normal ventricular function at the time of their combined operations, whereas 170 others had either segmental or diffuse ventricular impairment. Cumulative 5 year survival rates of 81% and 83% for each of these two subsets were nearly identical (Fig. 4, A). Fifty-eight patients in the follow-up group had single coronary grafts, 125 had double grafts, and 129 had three or more grafts. Five-year actuarial survival for each of these subsets was 66%, 87%, and 92%, respectively (Fig. 4, B). Cumulative survival among patients receiving single grafts, the majority of whom had severe CAD but only limited targets for revascularization, was significantly worse than among those who had double (p < 0.005) or multiple (p < 0.01) coronary grafts. Of the 312 operative survivors, 22 had single-vessel coronary atherosclerosis, 96 had double- vessel disease, and 194 had triple-vessel involvement. Sixty-seven of those with multiple-vessel CAD had significant stenosis of the left main coronary artery preoperatively. Cumulative 5 year survival rates for these subsets was 73%, 66%, 83%, and 80%, respectively (Fig. 5, A). Complete revascularization of all coronary lesions exceeding 50% luminal stenosis had been performed in 185 patients, while 127 others had had only incomplete revascularization. Actuarial survival 5 years postopera-
tively for these two subsets was 85% and 80% (Fig. 5, B). None of these differences was statistically valid.
Neurologic status and carotid disease. One hundred forty-six of the 312 patients in the follow-up group had experienced neurologic symptoms preceding their combined operations, whereas the remaining 166 had asymptomatic carotid stenosis. Five-year cumulative survival rates for these two subsets were 80% and 85%, respectively (Fig. 6, A). One hundred ninety patients had unilateral carotid stenosis at the time of their combined operations, and bilateral carotid disease exceeding 50% stenosis had been documented in 122. Fiveyear actuarial survival rates were 86% for patients with unilateral carotid disease and 78% for those having bilateral carotid stenosis (Fig. 6, B). None of these differences was statistically significant. Late neurologic complications. Eighteen (5.8%) of the 312 patients in the follow-up group have experienced late strokes at intervals of 4 to 85 months postoperatively (mean 30 months). Limited clinical information was available for three of these patients. In the remaining 15, late strokes appeared to involve the cerebral cortex on the same side as combined carotid endarterectomy in five (1.6%), the contralateral cerebral hemishpere in seven (2.4%), and the brain stem in three (1. 0%). Four of these strokes were fatal. Ten (4.9%) of 203 patients with less than 50% stenosis of the contralateral internal carotid artery at the
Volume 85
Coronary bypass and carotid endarterectomy
Number 4
585
April,1983
..::---0- _
100 """=---0
-.
90 A.
....
c(
>
s
II<
~
'"
60
o
NORMAL VENTRICLE (N=142)
• SEGMENTAL OR DIFFUSE IMPAIRMENT (N=170)
0
2
100 r-==--1I
• -=:::::::::::::=--.2
90 B.
.~.
80 70
~
~::::::.:::::o
___
80 70 60
4
3
1I_ _
1I_=----."
._.~.-------0
• SINGLE CABG (N=S8)
o DOUBLE CABG
5
__•
(N=12S)
"3 OR MORE CABG (N=129)
0
2
3
4
5
POSTOPERATIVE YEARS
Fig. 4. Life-table survival curves for (A) patients with normal ventricular function and those with segmental or diffuse ventricular impairment and (B) patients having one, two, and three or more coronary artery bypass grafts (CABG).
time of combined carotid endarterectomy have had late strokes. In comparison, late strokes have occurred in six (8.3%) of 72 patients who had over 50% stenosis of the contralateral internal carotid artery and in two (5.4%) of 37 patients with documented occlusion of the contralateral internal carotid artery. These differences, however, were not statistically significant. Additional cardiovascular operations. Fifty-nine patients (19%) have required a total of 69 elective peripheral vascular operations subsequent to combined carotid endarterectomy and myocardial revascularization. Contralateral carotid reconstruction has been performed in 30 patients. Twelve of these patients had planned, staged procedures to correct contralateral carotid disease that had been demonstrated at the time of their initial cerebral angiographic study, whereas 18 incidentally developed new neurologic sypmtoms or contralateral carotid stenosis during the late postoperative interval. Three other patients have required reoperations because of recurrent carotid stenosis. The remaining 36 patients have undergone elective aortic aneurysm resection (II), aortoiliac (13) or femoropopliteal (seven) reconstruction, renal revascularization (three), profundaplasty (one), or lumbar sympathectomy (one). Two postoperative deaths occurred, for a patient mortality of 3.4% and a procedure mortality of 2.9%.
Discussion One of the fundamental issues that must be addressed in any analysis of the surgical management of patients having severe, simultaneous coronary and carotid artery disease is whether staged or combined carotid reconstruction effectively reduces the risk of perioperative stroke among patients scheduled for elective myocardial revascularization. Although several previous publications have indicated an instinctive preference for the combined approach in this situation.v '! Liebman, and Marszalek'" and Turnipseed, Berkoff, and Belzer" have suggested that suspected carotid disease is present in 12% to 14% of such patients but does not represent a measurable threat for postoperative stroke after aorta-coronary bypass or any other cardiovascular procedure. Unfortunately, the conclusion that associated carotid disease may be disregarded in patients with coronary atherosclerosis was based in both of these reports upon the results of noninvasive cerebrovascular testing without angiographic confirmation. Since we have documented severe carotid atherosclerosis with the liberal use of angiography in only 1.7% of a large series of cardiac patients, the accuracy of the noninvasive studies employed in these and similar investigations is open to speculation. Moreover, it is distinctly difficult to quantitate the precise early benefit of carotid endarterectomy at the time of myocardial revas-
The Journal of
586
Hertzer et al.
Thoracic and Cardiovascular Surgery
100 90 A.
80 70
~
;(
60
>
0
'"
100
:> j
'"
•
SINGLE VESSEL DISEASE (N=22) DOUBLE VESSEL DISEASE (N=96) Do TRIPLE VESSEL DISEASE (N=194) x LEFT MAIN CORONARY DISEASE (N=67)
o
---------
-0-____
90 B.
60 0
4
_
COMPLETE CORONARY REVASCULARIZATION (N=lBS)
o
INCOMPLETE CORONARY REVASCULARIZATION (N=127)
2
3
5
-____
~--o
80 70
----3
2
-0
4
POSTOPERATIVE YEARS
Fig. 5. Life-table survival curves for (A) patients with single-vessel, double-vessel, triple-vessel, and left main coronary atherosclerosis and (B) patients having complete or incomplete myocardial revascularization.
cularization among the relatively few patients having severe carotid stenosis proved by angiography. For example, the incidence of perioperative stroke in the series of 22,100 patients undergoing aorta-coronary bypass grafting during the period of our study would have increased only from 1.8% to 2.7% even if known carotid disease had been ignored and one of every two patients in the study group had had completed strokes as the consequence. The results of this investigation indicate that combined myocardial revascularization and carotid endarterectomy may be performed with a reasonable margin of safety in patients whose coronary and carotid disease is considered too severe to permit staged operations. In our experience, the mortality (5.7%) and the incidence of permanent postoperative stroke (4.5%) after combined procedures have exceeded those expected for patients without carotid disease who require only aortacoronary bypass. However, associated extracranial disease was not the only additional risk factor identified in the study group. Mean age and the prevalence of left main coronary stenosis, multiple-vessel CAD, and impaired ventricular function in these patients were significantly higher than those found in the majority of patients undergoing myocardial revascularization alone. Moreover, statistical data do not entirely reflect the desperate clinical situation of many of these patients
who had incapacitating angina pectoris and diffuse CAD, compounded by impending stroke or severe carotid stenosis that often was bilateral. Our approach of surgical management for severe carotid disease demonstrated by angiography does not resolve the controversy concerning whether carotid reconstruction reduces the risk for stroke during myocardial revascularization, even though this risk was likely to be heavily concentrated among those selected for our study group. It appears that this issue will eventually require a large, prospectively randomized, and probably multicentered study using strict criteria that should include preoperative cerebral angiography. At the present time, we continue to recommend combined coronary and carotid procedures to patients with recognized carotid disease who require myocardial revascularization for (1) severe lesions of the left main coronary artery, (2) diffuse CAD without satisfactory collateral circulation, and (3) severe CAD and any form of unstable angina pectoris. Late results in patients with severe, simultaneous coronary and carotid artery disease represent a second fundamental consideration that too often is not addressed by studies that are limited to early postoperative complications. A number of reports have confirmed that carotid endarterectomy measurably reduces the incidence of late stroke among patients with previ-
Volume 85
Coronary bypass and carotid endarterectomy
Number 4
5 87
April, 1983
100 90 A.
~
:;( >
s
'" ::;)
80 70
o ASYMPTOMATIC
60
• PREOPERATIVE NEUROLOGIC SYMPTOMS (N=146)
0
90 B.
-0-
_
5
4
:3
2
100 r-:::::----o_
II>
CAROTID STENOSIS (N=166)
0 ____
-_____
0 __
------_________
0
-
80 70 60 0
c UNILATERAL CAROTID STENOSIS (N=190) _ BILATERAL CAROTID STENOSIS (N=122)
2
5
4
:3
POSTOPERATlVE YEARS
Fig. 6. Life-table survival curves for (A) patients with asymptomatic carotid stenosis and those having preoperative neurologic symptoms and (B) patients with unilateral carotid stenosis and those with bilateral carotid disease.
ous neurologic symptoms caused by extracranial arterial disease. 16 - 18 • 20 Moreover, Thompson, Patman, and Talkington'? and Cooperman, Martin, and Evans'" have presented convincing evidence that late neurologic complications occur with significant frequency in patients with asymptomatic carotid stenosis unless carotid endarterectomy is performed. Clearly, advanced carotid atherosclerosis discovered in patients with CAD should not simply be disregarded, irrespective of whether it is considered to represent a risk during myocardial revascularization. Symptomatic carotid lesions or asymptomatic stenosis should be evaluated and treated on their own merit even if staged operations after aorta-coronary bypass are preferred. Considering the concentration of severe extracranial disease among the patients in our study group, the fact that only 1.6% had late strokes involving the ipsilateral cerebral hemisphere during a mean postoperative interval of 38 months probably represents a substantial improvement over the results that could have been anticipated if carotid endarterectomy had not been performed. Although patients with severe carotid atherosclerosis comprise a relatively small subset of all those who require myocardial revascularization, the late mortality caused by myocardial infarction after carotid reconstruction suggests that CAD is prevalent among patients with carotid disease. 13-19 Table VI contains ac-
Table VI. Comparison of cumulative survival after successful extracranial arterial reconstruction Cumulative survival
I
I
5 yr
Series
Patients
I yr
DeBakey et al. II; (1965) DeWeese et al. 17 (1973) Bouchard et al. I.; (1975) Cleveland Clinic':' (1981) Cleveland Clinic (present series)
761
0.913
0.821
0.723
102
0.900
0.770
0.660
227
0.911
0.835
0.777
325
0.941
0.831
0.726
312
0.965
0.923
0.829
3 yr
tuarial data calculated for the 312 patients in our follow-up group in comparison to similar information for all operative survivors after extracranial arterial reconstruction presented in other published series;": 15-17 Cumulative survival at 1, 3, and 5 years for patients with recognized CAD who underwent myocardial revascularization as well as carotid endarterectomy is consistently superior to that for patients whose preoperative evaluations generally were restricted to their cerebrovascular circulation. Moreover, these differences may be expected to become even more obvious in
588
Hertzer et al.
the future, as late survival after myocardial revascularization continues to improve because of technical advances and contemporary attempts to provide complete revascularization with multiple aorta-coronary grafts.": 30. 31 From this perspective, it appears that recognition and management of severe, surgically correctable CAD in patients with known carotid disease is as important as the diagnosis of carotid atherosclerosis among those scheduled for myocardial revascularization. Neither may safely be ignored. Mrs. Ramona Zurrer and Mr. Eric Christiansen, Research Associates, coordinated the retrieval of late postoperative data. Richard Greenstreet, Ph.D., Department of Biostatistics, and Miss Geri Locher, Research Division, provided the statistical analysis. Miss Linda Maljovec assisted in preparation of the manuscript.
2
3
4
5
6
7
8
9
10
REFERENCES Ashor GW, Meyer BW, Lindesmith GG, Stiles QR, Walker GH, Tucker BL: Coronary artery disease. Surgery in 100 patients 65 years of age and older. Arch Surg 107:30-33, 1973 Brewer AC, Furlan AJ, Hanson MR, Lederman RJ, Loop FD, Cosgrove DM, Ghattas MA, Estafanous FG: Neurologic complications of open heart surgery. Computer-assisted analysis of 531 patients. Cleve Clin Q 48:205-206, 1981 Loop FD, Cosgrove DM, Lytle BW, Thurer RL, Simpfendorfer C, Taylor PC, Proudfit WL: An II year evolution of coronary arterial surgery (1967-1978). Ann Surg 190:444-455, 1979 Reul FJ, Morris GC Jr, Howell JF, Crawford ES, Stelter WJ: Current concepts in coronary artery surgery. A critical analysis of 1,287 patients. Ann Thorac Surg 14:243259, 1972 Bernhard VM, Johnson WD, Peterson 11: Carotid artery stenosis. Association with surgery for coronary artery disease. Arch Surg 105:837-840, 1972 Hertzer NR, Loop FD, Taylor PC, Beven EG: Staged and combined surgical approach to simultaneous carotid and coronary vascular disease. Surgery 84:803-811, 1978 Mehigan JT, Buch WS, Pipkin RD, Fogarty TJ: A planned approach to coexistent cerebrovascular disease in coronary artery bypass candidates. Arch Surg 112: 14031409, 1977 Morris GC, Ennix CL Jr, Lawrie GM, Crawford ES, Howell JF: Management of coexistent carotid and coronary artery occlusive atherosclerosis. Cleve Clin Q 45: 125-127, 1977 Okies JE, Macmanus Q, Starr A: Myocardial revascularization and carotid endarterectomy. A combined approach. Ann Thorac Surg 23:560-563, 1977 Rice PL, Pifarre R, Sullivan JH, Montoya A, Bakhos M: Experience with simultaneous myocardial revascularization and carotid endarterectomy. J THoRAc CARDIOVASC SURG 79:922-925, 1980
The Journal of Thoracic and Cardiovascular Surgery
II Urschel HC, Razzuk MA, Gardner MA: Management of concomitant occlusive disease of the carotid and coronary arteries. J THORAC CARDIOVASC SURG 72:829-834, 1976 12 Ennix CL Jr, Lawrie GM, Morris GC Jr, Crawford ES, Howell JF, Reardon MJ, Weatherford SC: Improved results of carotid endarterectomy in patients with symptomatic coronary disease. An analysis of 1,546 consecutive carotid operations. Stroke 10:122-125,1979 13 Hertzer NR, Lees CD: Fatal myocardial infarction following carotid endarterectomy. Three hundred thirty-five 'patients followed 6-11 years after operation. Ann Surg 194:212-218, 1981 14 Bauer RB, Meyer JS, Fields WS, Remington R, MacDonald MC, Callen P: Joint study of extracranial arterial occlusion. III. Progress report of controlled study of long-term survival in patients with and without operation. JAMA 208:509-518, 1969 15 Bouchard JP, Fabia J, Simard D, Drolet M, Cote J, Roy P: Carotid endarterectomy. Survival of 227 patients. Can Med Assoc J 113:949-951, 1975 16 DeBakey ME, Crawford ES, Cooley DA, Morris GC Jr, Garrett HE, Fields WS: Cerebral arterial insufficiency. One to II-year results following arterial reconstructive operation. Ann Surg 161:921-945, 1965 17 DeWeese JA, Rob CG, Satran R, March DO, Joynt JF, Summers D, Nicholas C: Results of carotid endarterectomies for transient ischemic attacks-five years later. Ann Surg 178:258-264, 1973 18 Thompson JE, Austin DJ, Patman RD: Carotid endarterectomy for cerebrovascular insufficiency. Long-term results in 592 patients followed up to thirteen years. Ann Surg 172:663-679, 1970 19 Thompson JE, Patman RD, Talkington CM: Asymptomatic carotid bruit. Long-term outcome of patients having endarterectomy compared with unoperated controls. Ann Surg 188:308-316, 1978 20 Lees CD, Hertzer NR: Postoperative stroke and late neurologic complications after carotid endarterectomy Arch Surg 116:1561-1568, 1981 21 Hertzer NR, Young JR, Kramer JR, Phillips DF, deWolfe VG, Ruschhaupt WF III, Beven EG: Routine coronary angiography prior to elective aortic reconstruction. Arch Surg 114: 1336-1344, 1979 22 Hertzer NR, Santoscoy TG, Langston RHS: Accuracy of carotid compression tonography in the diagnosis of carotid artery stenosis. Correlation with arteriography in 300 patients. Cleve Clin Q 47:79-87, 1980 23 Loop FD: Saphenous vein bypass graft, Modem Technics in Surgery: Cardiac/Thoracic Surgery, LH Cohn, ed., Mount Kisco, New York, 1979, Futura Publishing Company, Chap 10, pp 1-10 24 Beven EG: Carotid endarterectomy. Surg Clin North Am 55: 1111-1124, 1975 25 Mantel N: Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 50: 163-170, 1966 26 National Center for Health Statistics: Vital Statistics of the
Volume 85 Number4 April, 1983
United States (1976), Vol II-Mortality, Part A, Table 5 27 Barnes RW, Liebman PR, Marszalek BP: The natural history of asymptomatic carotid disease in patients undergoing cardiovascular surgery. Surgery 90: 1075-1082, 1981 28 Turnipseed WD, Berkoff HA, Belzer FO: Postoperative stroke in cardiac and peripheral vascular disease. Ann Surg 192:365-368, 1980 29 Cooperman M, Martin EW Jr, Evans WE: Significance of asymptomatic carotid bruits. Arch Surg 113: 1339-1340, 1978
Coronary bypass and carotid endarterectomy
589
30 Jones EL, Craver JM, King SB III, Douglas JS, Bradford JM, Brown M, Bone OK, Hatcher DR Jr: Clinical, anatomic and functional descriptors influencing morbidity, survival and adequacy of revascularization following coronary bypass. Ann Surg 192:390-402, 1980 31 Rahimtoola SH, Grunkemeier GL, Teply JF, Lambert LE, Thomas DR, Suen YF, Staff A: Changes in coronary bypass surgery leading to improved survival. JAMA 246:1912-1916,1981
CARDIOVASC SURG
85:577-589, 1983
Combined myocardial revascularization and carotid endarterectomy Operative and late results in 331 patients Simultaneous coronary artery bypass and carotid endarterectomy were performed in 331 patients (mean age 61 years) at the Cleveland Clinic from 1973 through 1981, Of these, 195 (59%) had Functional Class lll-IV angina pectoris, 308 (93%) had multiple-vessel coronary artery disease (CAD), 68 (21%) had over 50% stenosis of the left main coronary artery, and 185 (56%) had either segmental or diffuse impairment of left ventricular function, Asymptomatic carotid stenosis was documented in 173 patients (52'7c), and the remaining 158 had experienced either previous transient cerebral ischemia (38%) or completed strokes (10%). Single aorta-coronary grafts were placed in 59 patients (18%), double gratis in 131 (40%), and three or more grafts in 141 (42%). Nineteen patients (5.7%) died postoperatively in the hospital, Neurologic deficits occurred in 30 patients (9.0%) and produced permanent functional impairment in 15 (4.5%). Late results have been obtained for 312 operative survivors at a mean postoperative interval of 38 months. Thirty-eight patients (12%) have died, but the 5 year life-table survival rate of the study group was identical to that of the normal population aged 61 years, Significant differences in cumulative 5 year survival rates were identified among diabetic patients (p < 0.025) and among those receiving single rather than double (p < .005) or multiple (p < .01) coronary grafts. Although 18 patients (5.8%) have had late strokes, only five (1.6%) of these strokes have involved the cerebral hemisphere on the same side as combined carotid endarterectomy,
Norman R, Hertzer, M,D., Floyd D. Loop, M.D., Paul C. Taylor, M.D., and Edwin G. Beven, M.D., Cleveland, Ohio
Myocardial infarction and stroke are the two leading causes of cardiovascular death in the United States. It is not surprising, therefore, that severe, simultaneous coronary and carotid artery disease may be documented in some patients initially under consideration for either elective myocardial revascularization or extracranial arterial reconstruction alone. Although multifocal atherosclerosis is common to all such patients, their referral .for surgical management usually assumes one of two distinctly different courses. First, a relatively small subset of those with known coronary artery disease (CAD) is found to have adFrom the Cleveland Clinic Foundation, The Department of Thoracic and Cardiovascular Surgery and The Department of Vascular Surgery, Cleveland, Ohio. Received for publication May 24, 1982. Accepted for publication July 2, 1982. Address for reprints: Norman R. Hertzer, M.D., Department of Vascular Surgery, The Cleveland Clinic Foundation, 9500 Euclid Ave., Cleveland, Ohio 44106.
vanced carotid artery lesions preceding elective myocardial revascularization. The incidence of postoperative stroke after aorta-coronary bypass using contemporary methods for extracorporeal perfusion usually does not exceed 2%,1-4 and many of these neurologic deficits are manifest as diffuse encephalopathy or intellectual impairment that does not specifically implicate extracranial occlusive disease. Nevertheless, hemispheric neurologic complications might be expected to occur with greater frequency among patients with recognized carotid stenosis, and several previous reports have recommended that carotid endarterectomy should be performed in selected patients as a staged or combined procedure at the time of myocardial revascularization :;-11 Second, coronary atherosclerosis is prevalent among patients who have carotid disease that is sufficiently severe to warrant surgical correction. Provided the operative stroke rate is not excessive, myocardial infarctions are principally responsible for the few postoperative deaths that occur after carotid endarterectomy.P: 13 577
The Journal of Thoracic and Cardiovascular
578 Hertzer et al.
Surgery
Table I. Extent and distribution of coronary artery disease in 331 patients Coronary artery disease >50%sterwsis Distribution
No.
Left coronary artery Main trunk Anterior descending Circumflex Right coronary artery Single vessel Double vessel Triple vessel
68 286 236 273 23 98 210
I
> 70% stenosis
%
No.
21 86 71 82 7 30 63
39 244 191 245 51 131 135
I
%
12 74 58 74 16 40 40
Moreover, myocardial infarctions account for as many as 70% of all late deaths following successful extracranial arterial reconstruction and are especially common among patients suspected to have CAD on clinical grounds. 13-19 The incidence of fatal myocardial infarction during the late postoperative interval after carotid endarterectomy at the Cleveland Clinic has been shown to equal or surpass the total incidence of both fatal and nonfatal stroke in all age groups.s? In an attempt to identify those patients with cerebrovascular disease who have severe, associated CAD, coronary angiography has been routinely employed before elective carotid endarterectomy at this institution since 1978. 21 In a previous report, we described early postoperative results for the surgical management of simultaneous carotid and coronary vascular disease utilizing staged operations in 59 patients and a combined approach in 115 others. 6 The present investigation is an analysis of intermediate-term results as well as postoperative complications for a series of 331 patients who required combined myocardial revascularization and carotid endarterectomy at the Cleveland Clinic from 1973 through 1981.
Patients and metbods The 331 patients in this series represent only 1.7% of 22,100 patients who had direct myocardial revascularization at the Cleveland Clinic during the period of this study. There were 259 men (78%) and 72 women (22%), ranging in age from 39 to 85 years (mean 61 years). Nineteen (6%) were less than 50 years of age, 108 (32%) were 50 to 59 years of age, 145 (44%) were 60 to 69 years of age, and 59 (18%) were over 70 years of age. In comparison, the mean age of all 22,100 patients undergoing aorta-coronary bypass from 1973 through 1981 was 54 years.
Atherosclerotic risk factors. Diabetes mellitus, defined as known diabetes under medical treatment or disclosed by routine preoperative studies (abnormal glucose tolerance test, fasting blood sugar > 140 mg/dl) was present in 79 patients (24%). One hundred sixty-three patients (49%) had hypertension under medical management or had sustained blood pressure measurements above 150/90 mm Hg. One hundred seventyeight patients (54%) either smoked cigarettes or had discontinued their chronic use of tobacco within the previous 5 years. Serum cholesterol levels ranged from 95 to 541 mg/dl (mean 267 mg/dl), and serum triglyceride levels ranged from 51 to I ,223mg/dl (mean 216 mg/dl).· Patient selection. Patients scheduled for myocardial revascularization during this study period were questioned specifically concerning possible symptoms of previous completed strokes or transient cerebral ischemic attacks (TIA). Auscultation for carotid bruits was a routine feature of physical examination. Carotid angiography was performed for patients with previous neurologic symptoms and for those with asymptomatic cervical bruits that appeared to originate at the carotid bifurcation, particularly when the results of noninvasive assessment of internal carotid blood flow were abnormal.P Documentation of carotid disease was obtained by the transfemoral (Seldinger) approach, by cineangiography at the time of cardiac catheterization, or by digital subtraction angiography for all patients in this series. Carotid endarterectomy was recommended for patients who had previous neurologic symptoms and appropriate carotid bifurcation disease, as well as for those having asymptomatic bruits caused by stenosis of the cervical segment of the internal carotid artery that reduced luminal diameter by 70% or more. Asymptomatic stenosis compromising the residual lumen by less than 70% was occasionally repaired in the presence of contralateral internal carotid occlusion. Preoperative cardiac status. Symptoms. As defined by the New York Heart Association, Functional Class II angina pectoris was present in 106 patients (32%), Class III in 132 (40%), and Class IV in 63 (19%). Thirty patients (9%) had no angina pectoris at the time of their initial investigations, but 32 others (10%) had experienced symptoms of congestive heart failure. Electrocardiography. A standard, twelve-lead electrocardiogram (ECG) was normal in 92 patients (28%). Eighty-three patients (25%) had Q waves confirming previous myocardial infarctions, and III others (33%) had ischemic ST-T segment changes. Five patients (2%) had other ECG abnormalities such as left bundle
Volume 85 Number 4 April, 1983
Coronary bypass and carotid endarterectomy
100 65
579
PATIENTS O ALLIN=22,100)
60 55
•
COMBINED GROUP (N=331)
50 45
....
...z u
......'"
40 35 30 25 20 15 10 5
CORONARY ARTERY DISEASE
VENTRICULAR FUNCTION
PERIOPERATIVE COMPLICATIONS
Fig. 1. Graphic representation of the prevalence of preoperative risk factors and the incidence of postoperative complications for the 331 patients in this report and for the total of 22,100 patients undergoing myocardial revascularization during this period of study.
branch block, and 40 (12%) had two or more ECG findings. Cardiac catheterization. Table I presents a summary of data obtained during coronary angiography for all patients in this report, according to whether the maximum degree of angiographic stenosis exceeded either 50% or 70% of luminal diameter. The left main coronary artery was more than 70% stenotic in 39 patients (12%) and the anterior descending branch in 244 (74%). The presence of high-grade, subtotal coronary artery stenosis dictated a combined operation rather than a staged approach to associated carotid disease in many of these patients. More than 50% stenosis was found in a single coronary artery in 23 patients (7%), two coronary arteries in 98 (30%), and three coronary arteries in 210 (63%). According to the same criterion of 50% luminal stenosis, 14% of the total of 22,100 patients undergoing direct myocardial revascularization during this study period had single-vessel coronary disease, 30% had double-vessel disease, and 56% had triple-vessel disease. Cine left ventriculography demonstrated normal function in 146 patients (44%). Segmental akinesia was found in 155 patients (46%), and 30 (9%) had diffuse left ventricular impairment. In comparison, 52% of all 22,100 patients had normal ventriculograms, and 48% had some element of ventricular dysfunction.
Graphic representation of the extent of CAD (> 50% luminal stenosis) and left ventricular impairment for the series of 331 patients in this report and for the group of 22,100 patients from whom they were selected is presented in Fig. 1. The greater prevalence of triple-vessel disease (p < 0.0l), of left main coronary involvement (p < 0.00l), and of ventricular impairment (p < 0.01) in patients requiring combined procedures was statistically significant.
Preoperative carotid status. Symptoms. A history of previous neurologic symptoms was obtained from 158 patients (48%). Of these, 124 (38%) had experienced prior hemispheric or vertebrobasilar TIA, and 34 (10%) had had completed strokes. The remaining 173 patients (52%) had cervical bruits and were found to have severe, asymptomatic carotid stenosis during cerebral angiography. Angiographic findings. A summary of the extent and distribution of carotid artery disease determined by angiography is given in Table II. Only five patients (2%) had less than 50% stenosis of the internal carotid artery that was repaired during combined procedures, and each had ulcerated atheromatous disease as well as previous neurologic symptoms that could have been caused by microemboli arising from such lesions. The contralateral internal carotid artery was less than 50% stenotic in 216 patients (65%). Fifteen of these patients
The Journal of
580 Hertzer et al,
Thoracic and Cardiovascular Surgery
Table II. Extent and distribution of carotid artery disease in 331 patients Distribution Ipsilateral Carotid artery disease Stenosis <50% 50%·90% >90% Occlusion
No.
5 245 81
0
I
Contralateral
%
No.
2 74 24
216 69 9
37
I
%
65 21 3 II
had undergone contralateral carotid endarterectomy as a remote or staged procedure. Forty-six other patients (14%), however, had over 90% stenosis or occlusion of the contralateral internal carotid artery. Surgical management. Early in the study period, median sternotomy customarily was performed as the initial procedure so that prompt cannulation for cardiopulmonary bypass could be done if myocardial function deteriorated during carotid endarterectomy. Because of advances in anesthetic techniques and the pharmacologic support of cardiac performance, carotid endarterectomy in recent years has been completed before sternotomy with substantial savings in operative blood loss. Since harvesting and preparation of the saphenous vein for aorta-coronary grafting have usually been done at the same time as carotid endarterectomy, combined procedures have extended operating time only 45 minutes to I hour beyond that required for myocardial revascularization alone. Complete hemostasis and closure of the cervical incision was always delayed until intraoperative anticoagulation had been reversed and sternotomy closure had been completed. Coronary artery procedures. The technical details of direct myocardial revascularization at the Cleveland Clinic during the period of this study have been presented elsewhere.P: 23 Cold potassium cardioplegia was introduced in 1978 and, except for some patients with single-vessel disease, has been used in almost all revascularization operations since that time. Fifty-nine (18%) of the 331 patients in the present series had single aorta-coronary grafts, 131 (40%) had double grafts, 113 (34%) had triple grafts, and 28 (8%) had four or more grafts. Complete revascularization of all significant coronary artery lesions was accomplished in 197 patients (60%). In comparison, 28% of the total of 22,100 patients undergoing direct myocardial revascularization during this study period had single grafts, 38% had double grafts, 25% had triple grafts, and 9% had four or more grafts. In addition to carotid endarterectomy, 16 patients in
this report (4.8%) required other combined procedures, including aortic or mitral valve replacement in 12, open mitral commissurotomy in one, ventricular aneurysmectomy in two, and resection of a symptomatic abdominal aortic aneurysm in one. Carotid artery procedure. Technical aspects of carotid endarterectomy at this institution also have previously been described.v' Every patient received systemic sodium heparin, 5,000 to 10,000 international units, prior to the application of carotid clamps. An indwelling carotid shunt was inserted in 274 patients (83%), and only three of 185 operations since 1978 have been performed without shunts. Perfusion of the ipsilateral internal carotid artery was rarely interrupted for longer than 3 to 4 minutes during any operation done with the use of a shunt. Tacking sutures were frequently employed to secure the distal intima of the internal carotid artery, but arteriotomy closure required venous patch angioplasty in fewer than 5% of patients. Data analysis. The Fisher exact test of the null hypothesis was employed for analysis of paired variables. Differences in life-table data were evaluated by the method described by Mantel."
Early results Postoperative mortality. Nineteen of the 331 patients died in the hospital during the postoperative period, for an operative mortality rate of 5.7%. Although five patients eventually had multisystem failure, the principal cause of postoperative death was acute myocardial infarction or left ventricular failure in nine patients (2.7%), acute stroke or respiratory insufficiency in three each (0.9%), sepsis in two (0.6%), and renal failure or ruptured abdominal aortic aneurysm in one each (0.3%). Age and sex. No patient less than 50 years of age died postoperatively. Fatal complications occurred in eight (7.4%) of those 50 to 58 years of age, in five (3.4%) of those 60 to 69 years of age, and in six (10%) of those over 70 years of age. Mortality rates were 6.3% for 127 patients less than 60 years of age and 5.4% for 204 patients over 60 years of age. This difference was not statistically significant. Although postoperative death was more common among men (6.9%) than among women (1.4%), this trend was not firmly supported by statistical analysis (0.05 < p < 0.1). Diabetes and hypertension. No meaningful differences in hospital mortality rates were identified between diabetic (7.6%) and nondiabetic patients (5.2%) or between patients with hypertension (6.7%) and those who were normotensive (4.8%). Preoperative cardiac status. Neither the severity of
Volume 85 Number 4
Coronary bypass and carotid endarterectomy
581
April,1983
preoperative cardiac symptoms nor the extent and distribution of CAD had a significant influence upon postoperative mortality. Although mortality rates for patients with Functional Class I-II and Class III-IV angina pectoris were 2.9% and 7.7%, respectively, this difference was not statistically valid (0.05 < P < 0.1). Likewise, there were no perceptible differences in postoperative mortality among patients having singlevessel, double-vessel, or triple-vessel CAD (p > 0.1). However, postoperative mortality was higher among patients with segmental (7. 1%) or diffuse (13%) ventricular impairment than in those having normal preoperative ventricular function (2.7%). Four postoperative deaths occurred among 146 patients (2.7%) who had normal left ventriculograms, whereas there were 15 deaths in a total of 185 patients (8. 1%) with ventricular dysfunction (p < 0.05). Surgical management. There was only one postoperative death (1.7%) in the subset of 59 patients who received a single aorta-coronary bypass graft, and this death was caused by an acute myocardial infarction in a patient with additional severe but surgically uncorrectable CAD. Fatal complications occurred in seven (5.3%) of 131 patients who had double grafts, in nine (8.0%) of 113 who had triple grafts, and in two (7.1 %) of 28 who had four or more grafts. Twelve (6.1 %) of 197 patients died after complete revascularization of all significant coronary lesions, and there were seven deaths in the subset of 134 patients (5.2%) who received only incomplete revascularization. None of these differences was statistically meaningful. Throughout the period of this study, 21 patients (6.3%) had perioperative myocardial infarctions documented by serum glutamic oxaloacetic transaminase levels greater than 100 international units or by new Q waves on the postoperative ECG. Eight (39%) of these infarctions were associated with a fatal outcome. Preoperative carotid status. Thirteen (8.2%) of 158 patients with previous neurologic symptoms and six (3.5%) of 173 patients with asymptomatic carotid stenosis died after combined operations. There were eight deaths among 115 patients (7.0%) who had more than 50% stenosis or occlusion of the contralateral internal carotid artery and 11 deaths among 216 patients (5.1%) with less than 50% contralateral carotid stenosis. None of these differences was statistically significant. Postoperative neurologic complications. Thirty (9.0%) of the 331 patients had some degree of perioperative neurologic deficit as a direct complication of their combined operations. Sixteen of these deficits (4.8%) were manifest by symptoms and signs appropriate to the ipsilateral cerebral hemisphere, 10 (3.0%) to the contralateral cerebral hemisphere, and four
Table III. Technical considerations and results during consecutive periods of study 1973-1977 (N = 146) Factors and results
No.
Coronary artery Single bypass graft Double grafts Three or more grafts Periop. myocardial infarction Operative mortality Carotid artery Intraop. shunting Postop. neurologic deficit
I
%
1978-1981 (N = 185) No.
I
%
35 67 44 13 8
24 46 30 8.9 5.5
24 64 97 8 II
13 35 52 4.3 5.9
90
62 12.3
182 12
98 6.6
18
Table IV. Principal causes of late death among 3 J2 operative survivors
Principal cause oflate death
No.
Percent of late deaths
Percent of operative survivors
5 5
40 10 8 5 5 3 3 13 13
4.8 1.3 1.0 0.6 0.6 0.3 0.3 1.6 1.6
38
100
12.1
Myocardial infarction Stroke Ruptured aortic aneurysm Congestive heart failure Malignant neoplasm Chronic renal disease Pulmonary embolism Other Unknown
15 4 3 2 2
Totals
I I
(1.2%) to the brain stem. A total of 15 deficits (4.5%) were temporary, including 13 that resolved within a few days postoperatively and two completed strokes requiring a maximum of 2 months for complete recovery. Fifteen other completed strokes (4.5%) produced some element of permanent functional disability. Seven of the permanent strokes were associated with only mild impairment. Eight others were severe, and three of these (0.9%) were fatal. None of the postoperative neurologic deficits occurred in any patient less than 50 years of age. There were nine deficits among those 50 to 59 years of age (8.3%), 13 among those 60 to 69 years of age (9.0%), and eight among those over 70 years of age (14%). Neurologic deficits occurred in 7. 1% of patients less than 60 years of age and in 10% of those over 60 years of age. None of these differences was statistically significant. Likewise, there was no statistically valid trend in the incidence of postoperative neurologic deficits between diabetic (15%) and nondiabetic patients (7.1 %), between hypertensive patients (9.2%) and those with-
The Journal of Thoracic and Cardiovascular Surgery
582 Hertzer et al.
Table V. Complete life-table survival data, including operative mortality
Postop. year
No. alive at beginning of the year
No. lost to follow-up during the year
No. alive observed for only part of the year
No. exposed to risk of dying during the year
No. dying during the year
Proportion dying during the year
Proportion surviving the year
Proportion surviving to end ofeach year
30 days I 2 3 4 5 6 7 8 9
331 312 245 198 153 100 82 41 18 5
0 0 0 0 0 0 0 0 0 0
0 57 41 42 45 14 36 21 13 5
331 283.5 224.5 177 130.5 93 64 30.5 11.5 2.5
19 10 6 3 8 4 5 2 0 0
0.057 0.035 0.027 0.017 0.061 0.043 0.078 0.066 0.000 0.000
0.943 0.965 0.973 0.983 0.939 0.957 0.922 0.934 1.000 1.000
0.943 0.910 0.885 0.870 0.817 0.782 0.721 0.673 0.673 0.673
out hypertension (8.9%), between patients having normal ventricles (7.5%) and those with impaired ventricular function (10%), or among those receiving one (4.3%), two (9.2%), or three or more (9.5%) aortacoronary bypass grafts. Indications for operation. Fourteen (8.1%) of 173 patients with asymptomatic carotid stenosis had postoperative neurologic deficits. Deficits occurred in 16 (10%) of 158 patients who had experienced previous neurologic symptoms, including nine deficits among those with prior TIA (7.3%) and seven among those with completed strokes (21%). Despite an apparent tendency for those with previous strokes to sustain a higher incidence of postoperative neurologic deficits, the number of patients with previous strokes was too small to support this trend by statistical analysis (0.05 < P < 0.1). Preoperative carotid status. No neurologic complications occurred in any of the five patients with previous neurologic symptoms who had less than 50% stenosis of the ipsilateral carotid artery. Twenty-five (10%) of the 245 patients with 50% to 90% stenosis of the ipsilateral internal carotid artery had postoperative neurologic deficits, as did five (6.2%) of 80 patients having over 90% ipsilateral stenosis. This difference was not significant. However, the status of the contralateral internal carotid artery imposed a meaningful difference in the incidence of postoperative neurologic complications. Deficits occurred in 6.9% of those with less than 50% stenosis of the contralateral internal carotid artery, in 8.7% of those with 50% to 90% stenosis, and in 20% of those with over 90% stenosis or occlusion of the contralateral internal carotid artery. The neurologic complication rate of patients with over 90% stenosis or internal carotid occlusion was significantly higher than that of patients with less than
50% stenosis of the contralateral internal carotid artery (p < 0.02). Carotid shunting. Early in the study period, 57 patients underwent combined carotid endarterectomy without the use of an intraoperative shunt. Postoperative neurologic deficits occurred in seven (12%) of these patients. Indwelling shunts have been employed in 274 procedures, and 23 (8.4%) of these patients experienced postoperative neurologic complications. This difference did not attain statistical significance. Trends in morbidity and mortality. The hospital mortality rate for all 22,100 patients who underwent myocardial revascularization from 1973 to 1981 was 1.9% (Fig. 1). Postoperative strokes occurred in 1.8%, and 2.8% experienced perioperative myocardial infarctions. The mortality rate for the 331 patients who had combined operations was 5.7%. Postoperative neurologic deficits occurred in 9.0%, of whom 4.5% had permanent deficits, and 6.3% had perioperative myocardial infarctions. All of these differences were highly significant (p < 0.001). Table III contains comparable data concerning surgical management and the incidence of major postoperative complications after combined procedures during two consecutive periods of study, 1973 to 1977 (146 patients) and 1978 to 1981 (185 patients). During these intervals, there were obvious trends favoring multiple aorta-coronary bypass grafts and the routine use of intraoperative carotid shunting. Together with advances in anesthetic methods, intraoperative myocardial protection, and the technology of postoperative care, these trends appear to be associated with substantial reduction in the incidence of perioperative myocardial infarction and neurologic deficit in the study group. This improvement has not yet achieved statistical significance, however, and the mortality rate for combined
Volume 85 Number 4 , April, 1983
Coronary bypass and carotid endarterectomy
583
100
90 80 70 ~
60
:c>
50
'"
40
s J '"
30 20
•
COMBINED CORONARY-CAROTID OPERATIVE SURVIVORS (N=312)
a
CAROTID ENDARTERECTOMY OPERATIVE SURVIVORS, 1969-1973 (N=142)
D. NORMAL MALE POPULATION (1976), AGE 61
10 2
3
4
5
POSTOPERATIVE YEARS
Fig. 2. Life-table survival curves for 312 patients in this report, for a normal male population of the same age,26 and for 142 patients with suspected but uncorrected coronary artery disease after carotid endarterectomy. 13
myocardial revascularization and carotid endarterectomy has remained relatively constant. Late results Late survival. Complete late information has been obtained for all 312 operative survivors at a mean postoperative interval of 38 months. Thirty-eight (12%) of these 312 patients have died at a mean follow-up interval of 33 months, including 26 (11%) of 241 men (mean 34 months) and 12 (17%) of71 women (mean 32 months). Table IV presents a summary of the principal causes of late death, insofar as dependable information was available from family members, referring physicians, or civil authorities. Cardiovascular diseases are known to have been responsible for 24 (63%) of the 38 deaths. Myocardial infarction has accounted for 40% of late deaths but may have been more common, since the cause of sudden death without other verification was considered as unknown. Complete life-table data, including operative mortality, for the 331 patients in this series are given in Table V. Annual late mortality has ranged from 1.7% to 7.8%, and the 5 year cumulative survival for 312 operative survivors presently is 83% ±6% (SD). Fig. 2 provides graphic representation of life-table survival data for the 312 operative survivors in this study, for a normal male population of the same age (61
years);" and for 142 patients (mean age 60 years) with suspected but undocumented CAD who survived carotid endarterectomy at the Cleveland Clinic from 1969 through 1973. 13 The late survival curve for patients who underwent combined myocardial revascularization and carotid endarterectomy in this report is identical to that of the normal population during the first 3 postoperative years. A slight difference in 5 year survival between these two groups is not statistically significant. Actuarial 5 year survival of patients with suspected CAD from the 1969 to 1973 series was 62% ± '13% (SD). The difference in 5 year survival between operative survivors in this group and those in the present series of combined operations was statistically valid (p < 0.0005). Diabetes and hypertension. The 312 patients in the follow-up group included 73 with diabetes. Life-table survival curves for these patients and for 239 others who did not have diabetes are shown in Fig. 3, A. Cumulative 5 year survival rates of 69% for diabetic and 87% for nondiabetic patients were statistically different (p < 0.025). At the time of their combined operations, 175 patients were defined as hypertensive, while 137 others did not have hypertension. Life-table survival rates for these two subsets were 84% and 81%, respectively (Fig. 3, B). This difference was not statistically significant.
The Journal of
584 Hertzer et al.
Thoracic and Cardiovascular Surgery
100 90 A.
0__
•
~--------o--o--
-.~.----o
80 70 60
~
• DIABETIC OPERATIVE SURVIVORS (N=73)
o NON-DIABETIC
•
OPERATIVE SURVIVORS (N=239)
~
C
0
'"
100
>
s
:::l
'" B.
90
4
5
-a~.
•
~~~--------.--
80 70 60 0
0-
0
4
5
• HYPERTENSIVE OPERATIVE SURVIVORS (N=17 S)
o NORMOTENSIVE OPERATIVE SURVIVORS (N=137)
2
3
POSTOPERATIVE YEARS
Fig. 3. Life-table survival curves for (A) diabetic patients and those without diabetes and (B) hypertensive patients and those without hypertension.
Ventricular function and myocardial revascularization, One hundred forty-two of the 312 patients in the follow-up group had normal ventricular function at the time of their combined operations, whereas 170 others had either segmental or diffuse ventricular impairment. Cumulative 5 year survival rates of 81% and 83% for each of these two subsets were nearly identical (Fig. 4, A). Fifty-eight patients in the follow-up group had single coronary grafts, 125 had double grafts, and 129 had three or more grafts. Five-year actuarial survival for each of these subsets was 66%, 87%, and 92%, respectively (Fig. 4, B). Cumulative survival among patients receiving single grafts, the majority of whom had severe CAD but only limited targets for revascularization, was significantly worse than among those who had double (p < 0.005) or multiple (p < 0.01) coronary grafts. Of the 312 operative survivors, 22 had single-vessel coronary atherosclerosis, 96 had double- vessel disease, and 194 had triple-vessel involvement. Sixty-seven of those with multiple-vessel CAD had significant stenosis of the left main coronary artery preoperatively. Cumulative 5 year survival rates for these subsets was 73%, 66%, 83%, and 80%, respectively (Fig. 5, A). Complete revascularization of all coronary lesions exceeding 50% luminal stenosis had been performed in 185 patients, while 127 others had had only incomplete revascularization. Actuarial survival 5 years postopera-
tively for these two subsets was 85% and 80% (Fig. 5, B). None of these differences was statistically valid.
Neurologic status and carotid disease. One hundred forty-six of the 312 patients in the follow-up group had experienced neurologic symptoms preceding their combined operations, whereas the remaining 166 had asymptomatic carotid stenosis. Five-year cumulative survival rates for these two subsets were 80% and 85%, respectively (Fig. 6, A). One hundred ninety patients had unilateral carotid stenosis at the time of their combined operations, and bilateral carotid disease exceeding 50% stenosis had been documented in 122. Fiveyear actuarial survival rates were 86% for patients with unilateral carotid disease and 78% for those having bilateral carotid stenosis (Fig. 6, B). None of these differences was statistically significant. Late neurologic complications. Eighteen (5.8%) of the 312 patients in the follow-up group have experienced late strokes at intervals of 4 to 85 months postoperatively (mean 30 months). Limited clinical information was available for three of these patients. In the remaining 15, late strokes appeared to involve the cerebral cortex on the same side as combined carotid endarterectomy in five (1.6%), the contralateral cerebral hemishpere in seven (2.4%), and the brain stem in three (1. 0%). Four of these strokes were fatal. Ten (4.9%) of 203 patients with less than 50% stenosis of the contralateral internal carotid artery at the
Volume 85
Coronary bypass and carotid endarterectomy
Number 4
585
April,1983
..::---0- _
100 """=---0
-.
90 A.
....
c(
>
s
II<
~
'"
60
o
NORMAL VENTRICLE (N=142)
• SEGMENTAL OR DIFFUSE IMPAIRMENT (N=170)
0
2
100 r-==--1I
• -=:::::::::::::=--.2
90 B.
.~.
80 70
~
~::::::.:::::o
___
80 70 60
4
3
1I_ _
1I_=----."
._.~.-------0
• SINGLE CABG (N=S8)
o DOUBLE CABG
5
__•
(N=12S)
"3 OR MORE CABG (N=129)
0
2
3
4
5
POSTOPERATIVE YEARS
Fig. 4. Life-table survival curves for (A) patients with normal ventricular function and those with segmental or diffuse ventricular impairment and (B) patients having one, two, and three or more coronary artery bypass grafts (CABG).
time of combined carotid endarterectomy have had late strokes. In comparison, late strokes have occurred in six (8.3%) of 72 patients who had over 50% stenosis of the contralateral internal carotid artery and in two (5.4%) of 37 patients with documented occlusion of the contralateral internal carotid artery. These differences, however, were not statistically significant. Additional cardiovascular operations. Fifty-nine patients (19%) have required a total of 69 elective peripheral vascular operations subsequent to combined carotid endarterectomy and myocardial revascularization. Contralateral carotid reconstruction has been performed in 30 patients. Twelve of these patients had planned, staged procedures to correct contralateral carotid disease that had been demonstrated at the time of their initial cerebral angiographic study, whereas 18 incidentally developed new neurologic sypmtoms or contralateral carotid stenosis during the late postoperative interval. Three other patients have required reoperations because of recurrent carotid stenosis. The remaining 36 patients have undergone elective aortic aneurysm resection (II), aortoiliac (13) or femoropopliteal (seven) reconstruction, renal revascularization (three), profundaplasty (one), or lumbar sympathectomy (one). Two postoperative deaths occurred, for a patient mortality of 3.4% and a procedure mortality of 2.9%.
Discussion One of the fundamental issues that must be addressed in any analysis of the surgical management of patients having severe, simultaneous coronary and carotid artery disease is whether staged or combined carotid reconstruction effectively reduces the risk of perioperative stroke among patients scheduled for elective myocardial revascularization. Although several previous publications have indicated an instinctive preference for the combined approach in this situation.v '! Liebman, and Marszalek'" and Turnipseed, Berkoff, and Belzer" have suggested that suspected carotid disease is present in 12% to 14% of such patients but does not represent a measurable threat for postoperative stroke after aorta-coronary bypass or any other cardiovascular procedure. Unfortunately, the conclusion that associated carotid disease may be disregarded in patients with coronary atherosclerosis was based in both of these reports upon the results of noninvasive cerebrovascular testing without angiographic confirmation. Since we have documented severe carotid atherosclerosis with the liberal use of angiography in only 1.7% of a large series of cardiac patients, the accuracy of the noninvasive studies employed in these and similar investigations is open to speculation. Moreover, it is distinctly difficult to quantitate the precise early benefit of carotid endarterectomy at the time of myocardial revas-
The Journal of
586
Hertzer et al.
Thoracic and Cardiovascular Surgery
100 90 A.
80 70
~
;(
60
>
0
'"
100
:> j
'"
•
SINGLE VESSEL DISEASE (N=22) DOUBLE VESSEL DISEASE (N=96) Do TRIPLE VESSEL DISEASE (N=194) x LEFT MAIN CORONARY DISEASE (N=67)
o
---------
-0-____
90 B.
60 0
4
_
COMPLETE CORONARY REVASCULARIZATION (N=lBS)
o
INCOMPLETE CORONARY REVASCULARIZATION (N=127)
2
3
5
-____
~--o
80 70
----3
2
-0
4
POSTOPERATIVE YEARS
Fig. 5. Life-table survival curves for (A) patients with single-vessel, double-vessel, triple-vessel, and left main coronary atherosclerosis and (B) patients having complete or incomplete myocardial revascularization.
cularization among the relatively few patients having severe carotid stenosis proved by angiography. For example, the incidence of perioperative stroke in the series of 22,100 patients undergoing aorta-coronary bypass grafting during the period of our study would have increased only from 1.8% to 2.7% even if known carotid disease had been ignored and one of every two patients in the study group had had completed strokes as the consequence. The results of this investigation indicate that combined myocardial revascularization and carotid endarterectomy may be performed with a reasonable margin of safety in patients whose coronary and carotid disease is considered too severe to permit staged operations. In our experience, the mortality (5.7%) and the incidence of permanent postoperative stroke (4.5%) after combined procedures have exceeded those expected for patients without carotid disease who require only aortacoronary bypass. However, associated extracranial disease was not the only additional risk factor identified in the study group. Mean age and the prevalence of left main coronary stenosis, multiple-vessel CAD, and impaired ventricular function in these patients were significantly higher than those found in the majority of patients undergoing myocardial revascularization alone. Moreover, statistical data do not entirely reflect the desperate clinical situation of many of these patients
who had incapacitating angina pectoris and diffuse CAD, compounded by impending stroke or severe carotid stenosis that often was bilateral. Our approach of surgical management for severe carotid disease demonstrated by angiography does not resolve the controversy concerning whether carotid reconstruction reduces the risk for stroke during myocardial revascularization, even though this risk was likely to be heavily concentrated among those selected for our study group. It appears that this issue will eventually require a large, prospectively randomized, and probably multicentered study using strict criteria that should include preoperative cerebral angiography. At the present time, we continue to recommend combined coronary and carotid procedures to patients with recognized carotid disease who require myocardial revascularization for (1) severe lesions of the left main coronary artery, (2) diffuse CAD without satisfactory collateral circulation, and (3) severe CAD and any form of unstable angina pectoris. Late results in patients with severe, simultaneous coronary and carotid artery disease represent a second fundamental consideration that too often is not addressed by studies that are limited to early postoperative complications. A number of reports have confirmed that carotid endarterectomy measurably reduces the incidence of late stroke among patients with previ-
Volume 85
Coronary bypass and carotid endarterectomy
Number 4
5 87
April, 1983
100 90 A.
~
:;( >
s
'" ::;)
80 70
o ASYMPTOMATIC
60
• PREOPERATIVE NEUROLOGIC SYMPTOMS (N=146)
0
90 B.
-0-
_
5
4
:3
2
100 r-:::::----o_
II>
CAROTID STENOSIS (N=166)
0 ____
-_____
0 __
------_________
0
-
80 70 60 0
c UNILATERAL CAROTID STENOSIS (N=190) _ BILATERAL CAROTID STENOSIS (N=122)
2
5
4
:3
POSTOPERATlVE YEARS
Fig. 6. Life-table survival curves for (A) patients with asymptomatic carotid stenosis and those having preoperative neurologic symptoms and (B) patients with unilateral carotid stenosis and those with bilateral carotid disease.
ous neurologic symptoms caused by extracranial arterial disease. 16 - 18 • 20 Moreover, Thompson, Patman, and Talkington'? and Cooperman, Martin, and Evans'" have presented convincing evidence that late neurologic complications occur with significant frequency in patients with asymptomatic carotid stenosis unless carotid endarterectomy is performed. Clearly, advanced carotid atherosclerosis discovered in patients with CAD should not simply be disregarded, irrespective of whether it is considered to represent a risk during myocardial revascularization. Symptomatic carotid lesions or asymptomatic stenosis should be evaluated and treated on their own merit even if staged operations after aorta-coronary bypass are preferred. Considering the concentration of severe extracranial disease among the patients in our study group, the fact that only 1.6% had late strokes involving the ipsilateral cerebral hemisphere during a mean postoperative interval of 38 months probably represents a substantial improvement over the results that could have been anticipated if carotid endarterectomy had not been performed. Although patients with severe carotid atherosclerosis comprise a relatively small subset of all those who require myocardial revascularization, the late mortality caused by myocardial infarction after carotid reconstruction suggests that CAD is prevalent among patients with carotid disease. 13-19 Table VI contains ac-
Table VI. Comparison of cumulative survival after successful extracranial arterial reconstruction Cumulative survival
I
I
5 yr
Series
Patients
I yr
DeBakey et al. II; (1965) DeWeese et al. 17 (1973) Bouchard et al. I.; (1975) Cleveland Clinic':' (1981) Cleveland Clinic (present series)
761
0.913
0.821
0.723
102
0.900
0.770
0.660
227
0.911
0.835
0.777
325
0.941
0.831
0.726
312
0.965
0.923
0.829
3 yr
tuarial data calculated for the 312 patients in our follow-up group in comparison to similar information for all operative survivors after extracranial arterial reconstruction presented in other published series;": 15-17 Cumulative survival at 1, 3, and 5 years for patients with recognized CAD who underwent myocardial revascularization as well as carotid endarterectomy is consistently superior to that for patients whose preoperative evaluations generally were restricted to their cerebrovascular circulation. Moreover, these differences may be expected to become even more obvious in
588
Hertzer et al.
the future, as late survival after myocardial revascularization continues to improve because of technical advances and contemporary attempts to provide complete revascularization with multiple aorta-coronary grafts.": 30. 31 From this perspective, it appears that recognition and management of severe, surgically correctable CAD in patients with known carotid disease is as important as the diagnosis of carotid atherosclerosis among those scheduled for myocardial revascularization. Neither may safely be ignored. Mrs. Ramona Zurrer and Mr. Eric Christiansen, Research Associates, coordinated the retrieval of late postoperative data. Richard Greenstreet, Ph.D., Department of Biostatistics, and Miss Geri Locher, Research Division, provided the statistical analysis. Miss Linda Maljovec assisted in preparation of the manuscript.
2
3
4
5
6
7
8
9
10
REFERENCES Ashor GW, Meyer BW, Lindesmith GG, Stiles QR, Walker GH, Tucker BL: Coronary artery disease. Surgery in 100 patients 65 years of age and older. Arch Surg 107:30-33, 1973 Brewer AC, Furlan AJ, Hanson MR, Lederman RJ, Loop FD, Cosgrove DM, Ghattas MA, Estafanous FG: Neurologic complications of open heart surgery. Computer-assisted analysis of 531 patients. Cleve Clin Q 48:205-206, 1981 Loop FD, Cosgrove DM, Lytle BW, Thurer RL, Simpfendorfer C, Taylor PC, Proudfit WL: An II year evolution of coronary arterial surgery (1967-1978). Ann Surg 190:444-455, 1979 Reul FJ, Morris GC Jr, Howell JF, Crawford ES, Stelter WJ: Current concepts in coronary artery surgery. A critical analysis of 1,287 patients. Ann Thorac Surg 14:243259, 1972 Bernhard VM, Johnson WD, Peterson 11: Carotid artery stenosis. Association with surgery for coronary artery disease. Arch Surg 105:837-840, 1972 Hertzer NR, Loop FD, Taylor PC, Beven EG: Staged and combined surgical approach to simultaneous carotid and coronary vascular disease. Surgery 84:803-811, 1978 Mehigan JT, Buch WS, Pipkin RD, Fogarty TJ: A planned approach to coexistent cerebrovascular disease in coronary artery bypass candidates. Arch Surg 112: 14031409, 1977 Morris GC, Ennix CL Jr, Lawrie GM, Crawford ES, Howell JF: Management of coexistent carotid and coronary artery occlusive atherosclerosis. Cleve Clin Q 45: 125-127, 1977 Okies JE, Macmanus Q, Starr A: Myocardial revascularization and carotid endarterectomy. A combined approach. Ann Thorac Surg 23:560-563, 1977 Rice PL, Pifarre R, Sullivan JH, Montoya A, Bakhos M: Experience with simultaneous myocardial revascularization and carotid endarterectomy. J THoRAc CARDIOVASC SURG 79:922-925, 1980
The Journal of Thoracic and Cardiovascular Surgery
II Urschel HC, Razzuk MA, Gardner MA: Management of concomitant occlusive disease of the carotid and coronary arteries. J THORAC CARDIOVASC SURG 72:829-834, 1976 12 Ennix CL Jr, Lawrie GM, Morris GC Jr, Crawford ES, Howell JF, Reardon MJ, Weatherford SC: Improved results of carotid endarterectomy in patients with symptomatic coronary disease. An analysis of 1,546 consecutive carotid operations. Stroke 10:122-125,1979 13 Hertzer NR, Lees CD: Fatal myocardial infarction following carotid endarterectomy. Three hundred thirty-five 'patients followed 6-11 years after operation. Ann Surg 194:212-218, 1981 14 Bauer RB, Meyer JS, Fields WS, Remington R, MacDonald MC, Callen P: Joint study of extracranial arterial occlusion. III. Progress report of controlled study of long-term survival in patients with and without operation. JAMA 208:509-518, 1969 15 Bouchard JP, Fabia J, Simard D, Drolet M, Cote J, Roy P: Carotid endarterectomy. Survival of 227 patients. Can Med Assoc J 113:949-951, 1975 16 DeBakey ME, Crawford ES, Cooley DA, Morris GC Jr, Garrett HE, Fields WS: Cerebral arterial insufficiency. One to II-year results following arterial reconstructive operation. Ann Surg 161:921-945, 1965 17 DeWeese JA, Rob CG, Satran R, March DO, Joynt JF, Summers D, Nicholas C: Results of carotid endarterectomies for transient ischemic attacks-five years later. Ann Surg 178:258-264, 1973 18 Thompson JE, Austin DJ, Patman RD: Carotid endarterectomy for cerebrovascular insufficiency. Long-term results in 592 patients followed up to thirteen years. Ann Surg 172:663-679, 1970 19 Thompson JE, Patman RD, Talkington CM: Asymptomatic carotid bruit. Long-term outcome of patients having endarterectomy compared with unoperated controls. Ann Surg 188:308-316, 1978 20 Lees CD, Hertzer NR: Postoperative stroke and late neurologic complications after carotid endarterectomy Arch Surg 116:1561-1568, 1981 21 Hertzer NR, Young JR, Kramer JR, Phillips DF, deWolfe VG, Ruschhaupt WF III, Beven EG: Routine coronary angiography prior to elective aortic reconstruction. Arch Surg 114: 1336-1344, 1979 22 Hertzer NR, Santoscoy TG, Langston RHS: Accuracy of carotid compression tonography in the diagnosis of carotid artery stenosis. Correlation with arteriography in 300 patients. Cleve Clin Q 47:79-87, 1980 23 Loop FD: Saphenous vein bypass graft, Modem Technics in Surgery: Cardiac/Thoracic Surgery, LH Cohn, ed., Mount Kisco, New York, 1979, Futura Publishing Company, Chap 10, pp 1-10 24 Beven EG: Carotid endarterectomy. Surg Clin North Am 55: 1111-1124, 1975 25 Mantel N: Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 50: 163-170, 1966 26 National Center for Health Statistics: Vital Statistics of the
Volume 85 Number4 April, 1983
United States (1976), Vol II-Mortality, Part A, Table 5 27 Barnes RW, Liebman PR, Marszalek BP: The natural history of asymptomatic carotid disease in patients undergoing cardiovascular surgery. Surgery 90: 1075-1082, 1981 28 Turnipseed WD, Berkoff HA, Belzer FO: Postoperative stroke in cardiac and peripheral vascular disease. Ann Surg 192:365-368, 1980 29 Cooperman M, Martin EW Jr, Evans WE: Significance of asymptomatic carotid bruits. Arch Surg 113: 1339-1340, 1978
Coronary bypass and carotid endarterectomy
589
30 Jones EL, Craver JM, King SB III, Douglas JS, Bradford JM, Brown M, Bone OK, Hatcher DR Jr: Clinical, anatomic and functional descriptors influencing morbidity, survival and adequacy of revascularization following coronary bypass. Ann Surg 192:390-402, 1980 31 Rahimtoola SH, Grunkemeier GL, Teply JF, Lambert LE, Thomas DR, Suen YF, Staff A: Changes in coronary bypass surgery leading to improved survival. JAMA 246:1912-1916,1981