Surgical Management in Patients With Coexistent Coronary and Cerebrovascular Disease

Surgical Management in Patients With Coexistent Coronary and Cerebrovascular Disease* Long-Term Results Tej K. Kaul, MD, FCCP; Barry L. Fields, MD; David A. Wyatt, MD; Christopher R. jones, MD; and Donald R. Kahn, MD Between January 1980 and December 1992, 3% (210/ 6,862) of our patients undergoing myocardial revascularization (CABG) had high grade (>80%) internal carotid stenosis (CS). One hundred seventy-five of these patients with complete follow up for a minimum of 18 months were studied. Bilateral internal CS was present in 60%, and 75% had other vascular lesions, mainly as peripheral vascular disease (PVD) of the lower limb (50.8%). All patients underwent CAE (carotid endarterectomy) followed by CABG under the same anesthesia. Peripheral vascular lesions, contralateral internal CS and recurrent (n=43) and progressive vascular lesions (n=50), were subsequently treated as staged procedures. Hospital mortality was 3.42%. By univariate analysis significant predictors of late mortality were congestive heart failure, COPD, PVD, postoperative myocardial infarction, postoperative stroke, and ischemic cardiomyopathy. Only the latter two were also significant by multivariate analysis. At 12 years, actuarial survival in the presence of these risk factors were 46% , 49%, 22%, 37%, 53%, and 27% respectively. All are significantly lower as compared with the corresponding subsets of patients with the risk factor absent. At 12 years, actuar-

ial survival for the entire series was 65%. Cumulative incidence of postoperative strokes was higher in patients with bilateral internal CS than in patients with unilateral internal CS (p<0.07) and in patients with neurologic symptoms than asymptomatic patients. At 12 years, actuarial freedom from all cardiac related events, postoperative stroke, and symptomatic PVD were 49%, 82%, and 76% respectively. After successful revascularization these patients should be carefully followed for recurrent and progressive vascular lesions. (Chest 1994; 106:1349-57)

The presence of significant multifocal vascular disease is an indicator of an advanced stage of atherosclerosis. 1•2 Coexisting with significant coronary artery disease (CAD), internal carotid stenosis (CS) is often a difficult combination for the timing of surgical intervention especially in the presence of bilateral internal CS, unstable CAD,or at times in the presence of other peripheral vascular lesions. A synchronous combined approach has been preferred in this situation to minimize the risk of the sequelae from the uncorrected lesion.l To reduce the risk of postoperative neurologic events, Bernhard et aP advocated carotid artery endarterectomy (CAE) before coronary artery bypass graft (CABG), but under the same anesthetic. This approach has been successfully used to treat coexistent disease of the carotid and coronary circulation, with an acceptable mortality and morbidity. 3-15 Most studies have described a

better survivaP·5.l 6.l 7 and a lower incidence of neurologic complications 11 .l 2•18.l 9 after the combined approach as opposed to staged intervention. Only a few reports are available describing the long-term follow-up of these patients.l 1•14.l 5 Coexistent peripheral vascular disease (PVD) in this subset of patients has been usually treated subsequently as a staged procedure.10•11 If there is concurrent PVD threatening the viability of the lower limb, then simultaneous myocardial revascularization and reconstruction may be performed. 15 In this series , we have studied the factors influencing the short- and the long-term prognosis in these subsets of patients.

*From the Department of Cardiac Surgery , Princeton-Baptist Medical Center, Birmingham, Ala. Manuscript Received Fel:i. 7, 1994, accepted April19, 1994. Reprint requests: Dr. Kaul, Suite 300, Cardiac Surgery, Princeton Baptist Medical Center, Birmingham, AL 35211

CABG=coronary artery bypass graft; CAD=coronary artery disease; CAE=carotid artery endarterectomy; CHF =congestive heart failure; CI=confidence intervals; COPD =chronic obstructive pulmonary disease; CS=carotid stenosis; IMA=internal mammary artery; LV=left ventricular; L VEF= left ventricular ejection fraction; MI =myocardial infarction; PVD=peripheral vascular disease; TIA= transient ischemic attack

Key words: coronary artery bypass; coronary artery disease; cerebrovascular disease

METHODS

Between January 1980 and December 1991 , 175 (3%) patients who underwent myocardial revascularization also had a significant internal carotid stenosis (luminar stenosis >80%). These patients underwent carotid endarterectomy for the dominant lesion followed by myocardial revascularization on cardiopulmonary bypass under the same anesthetic. Of these, 123 (75.4%) patients also had other coexistent vascular lesions (Table 1). Fifty-five of the 89 patients with PVD had severe ischemia of the lower limbs. Of these, 53 underwent staged surgical interventions within 3 CHEST /106 / 5/ NOVEMBER, 1994

1349

Table !-Coexistent Preoperative Vascular Lesions

CAD Triple-vessel disease Two-vessel disease Rec urrent CAD Left main-stem disease Renal arterial stenosis Subclavian arterial stenosis Significant PVD of the lower limb* Abdominal aortic aneurysm Aortoiliac disease Ileofemoral disease Femoropopliteal and tibial Nonsignificant PVD of the lower limbt Internal CS Right ~85 % stenosis Left ~85% stenosis Bilateral! Recurrent stenosis External carotid or bifurcation of common carotid Intracranial disease Bilateral vertebral stenosis Right vertebral stenosis Left vertebral stenosis Basilar artery occlusion Middle cerebral disease

n

%

159 16 29 33 6 3

90.85 9.16 16.9 18.2 3.4

55 9 6 9 31 34

31.42

36 34 105 13

20.57 19.4 60.0 17.4

17 17 3 3 6 3 2

9. 7 9.7

1.7

Table 2-Surgical Procedures Performed* n

Procedure CABG CABG CABG left IMA Redo CABG Redo CABG left IMA Unilateral CAE Right CAE Left CAE Sequential or staged bilateral CAE (bilateral >85% stenosis) Redo CAE Renal arterial stenosis Angioplasty Reconstruction Nephrectomy Subclavian arterial Stenosis angioplasty Peripheral vascular procedures Aortic bifurcation graft llcofemoral bypass Angioplasty for ileofemoral lesion Laser dilation of ileofemoral lesion Femoropopliteal bypass Redo femoropopliteal or tibial bypass Amputations below knee Toe

140 35 24 5 74 71 30

13 6 2 1 2 53 (30.28%) 5 3 9 1 16 6 2

*Rest pain, ischemic ulceration / gangrene, claudication limiting active life. tRadiologic evidence/occasional claudication but no serious limitation of activity. !30 patients with bilateral internal CS has bilateral ~80% lesions, remaining patients had ~80% ipsilateral and ~50 % contralateral stenosis.

*CABG=4.07 graft/ patient. All patien ts underwent CABG and CAE for restoration of the circulation of the dominant side. Sequential or staged bilateral CAE was performed for contralateral disease at a mean interval of 4.4 ± 1.4 mon after the initial CAE+CABG. Shunts were used in 30 patients (Javid shunt, 28; Inahara Pruitt shunt, 2; Argyle shunt, 1). Sequential or staged peripheral vascular procedures were performed within 6 mon of initial CAE+CABG.

weeks to 6 months after combined CAE and CABG. Thirty patients with bilateral critical disease underwent staged CAE for significant contralateral internal carotid stenosis as urgent (within 3 weeks) or elective (within 1 to 6 months) (Table 2). Preoperative details, including coronary artery risk factors, distribution of vascular lesions, details of the surgical procedures

FIGUilE l. Preoperative cardiac status. Mild impairment of LV function (LVEF 40 to 50%). Moderate impairment of LV function (LVEF >30 to 40%). Severe impairment of LV function (LVEF <30%).

performed, and preoperative cardiac and neurologic status are shown in Figure 1 and Tables 1 and 2. Mean age was 66.12 ± 0. 78 yrs (range, 41 to 82 years); 122 were men, 53 women. Hypertension was present in 114 patients (65.14%), diabetes in 50 (28.9%), current smoking in 65 (37%), and family history of CAD in 39 patients (22.7%). Seventy-seven (44.8%) patients had established chronic obstructive pulmonary disease (COPD). Triple-vessel significant CAD was present in 159 (90.85%) patients, two-vessel disease in 16 (9. 15%), recurrent CAD in 29 (16.9%), and left mainstem disease in 33 (18.2%) patients. Preoperative unstable angina was present in 61 (34.28%) patients, recent myocardial infarction in 22 (12.5%), old myocardial infarction (MI) in 33 (19.2%), and poor left ventricular (LV) function, LV ejection fraction (LVEF) less than 30% in 20 (11.6) patients. Left internal mammary artery (IMA) was used in 40 (23.1%) patients. Preoperatively, 132 (75%) were in New York Heart Association class 1 to 2, 23 (13.14%) in class 3, and 20 (11.7%) patients were in class 4. Bilateral internal CS was present in 105 patients (60%), and 70 (40%) had unilateral CS. Preoperatively, transient ischemic attacks (TIAs) were experienced by 49 patients, amaurosis fugax by 19, vertebrobasilar symptoms by 8, hemispheric symptoms by 22, and preoperative stroke by 15 (8.5%); 111 (63. 4%) remaining patients had asymptomatic carotid bruit but severe internal CS (;:::85% stenosis). One hundred thirty-four (76.57%) patients were primarily referred for anginal symptoms, congestive heart failure (CHF), or MI. Earlier in this series, exercise or stress test was performed in

1350

Surgical Management in Coronary and Cerebrovascular Disease (Kauf et aO

PREOPERATIVE CARDIAC STATUS UORUAL IV LVEF ~ 60"~

•

RECENT t.U < 30 DAYS

D

OLOI.tl

•

RECENT Ml OLOMI

~

RECENT Ml

m

OLOMI

1.1tl0 LV OYSFUNC TI O tl

1.100ERA rE LV OVSF"UNCT!OU

SEVERE LV OYSFUtiC TIOtl LVEF < 30"o CHF

ABLE ANGINA

RECENT Ml

(n=61) RECENT r.u n OLOt.H n =- 33

OLOMI

10

20

30

40

50

60

n

70

Table 3-Late Deaths

Cardiac Ischemic cardiomyopathy Myocardial infarction Stroke Respiratory complications Renal failure Cancer Car accident Unknown

n

%

19 16 3 8 4 1 5

46.3

19.5 9.7 2.4 12.2 2.4 4.8

2

Table 4-Univariate

Preoperative CHF COPD PVD RAS* Postoperative infarction Postoperative CHF +ischemic cardiomyopathy Postoperative stroke

x2

df

p

5.3 6.24 86.37 86.23 23.5 37.5

2 2 4 4 4 4

<0.05 <0.044 <0.0001 <0.0001 <0.0001 <0.0001

27.297

4

<0.0001

*RAS=renal arterial stenosis; df=degrees of freedom. patients with no neurologic symptoms or sequelae, but later a dipyridamole thallium scan was primarily used before coronary angiography. All patients undergoing CABG with neurologic symptoms, carotid bruit, and patients older than 65 years in the study underwent carotid Doppler study. All patients included in this series had coronary angiography and four-vessel cerebral arteriogram performed at or during this same hospital admission. In the patients needing IMA grafting, extracranial cervical vessels were studied preoperatively by angiography. Fifty-five patients had significant symptomatic PVD. Of these, 18 patients underwent urgent, 29 underwent elective, and 6 patients underwent reoperation for the recurrent PVD. All peripheral vascular procedures were performed as staged procedures after combined CAE and CABG. Patients with the recurrent or progressive vascular disease (CAD, internal carotid, or PVD ) were reinvestigated by noninvasive (thallium scan, Doppler duplex scan), or invasive (angiography) methods and were reoperated on if indicated. Follow-up was complete in all patients by June 1993. Patients were followed up for a total period of 9,392.6 months (792.7 years), the mean follow-up was54.9 months (4.5 years), and minimum follow-up was 18 months. Six patients with incomplete follow-up and 29 more patients with less than 18 months ' follow-up have not -been included in this series. This period is desirable to assess full benefit from CAE. These patients had no neurologic deficits or MI following combined CAE and myocardial revascularization.

Statistical Analysis: Values are mean + SEM. Paired data were compared using Student's t test. Predictors of postoperative events were analyzed

by univariate analysis using x2 test. Factors with p value less than 0.1 were entered for stepwise multiple regression analysis. The following factors were entered as predictors of postoperative late deaths: age, sex, CAD risk factors, left mainstem disease, preoperative Ml, unstable angina, poor LV function (LVEF :$30%), CHF, COPD, unilateral carotid stenosis, bilateral carotid stenosis, PVD, renal arterial stenosis, reoperations for recurrent CAD, use of left IMA as conduit, preoperative stroke, postoperative ischemic cardiomyopathy and CHF, postoperative stroke, and the year of the operation. Coronary risk factors were entered for the prediction of vascular lesions and more advanced disease (bilateral internal CS and PVD). The following factors were entered for the prediction of postoperative neurologic events: age, sex, hypertension, diabetes, poor LV function , CHF, unilateral CS, bilateral CS, preoperative TIA, and preoperative strokes. Cox proportional hazard model was used for the comparison of the postoperative events (risk factor: present vs absent). Actuarial analysis was used to represent postoperative survival and events. The actuarial curves were compared using the test of Mantel 20 and Haenszel. A statistical analysis package (Stat pac) was used for the statistical analysis and p less than 0.05 was considered as significant. RESULTS

Postoperative Survival Thirty-day and hospital mortality was 3.42% (6/ 175) . Two early deaths were due to postoperative stroke (1 hemispheric, 1 brainstem death) . Four early deaths were cardiac related (perioperative MI and unstable angina [n=3]; CHF in a patient with LVEF less than 20% [n=1]) . The latter patient had also received an LV assist device after a trial period on an intra-aortic balloon pump. There were no deaths in the other high-risk patients who required intra-aortic balloon pump support (n=5), patients receiving hemodialysis for chronic or acute renal failure (n =4), and in patients who underwent balloon angioplasty for severe bilateral renal arterial stenosis (n=3). Details of late death are outlined in Table 3. Nineteen late deaths (3 months to 10 years) were due to the terminal stage of ischemic heart disease and 8 were due to postoperative stroke. On univariate analysis, the following risk factors had a significant influence over late mortality: preoperative CHF (p<0.05), COPD (p<0.044), PVD (p<0.001), postoperative MI (p<0.05), postoperative stroke (p<0.001), and postoperative ischemic cardiomyopathy with CHF (p<0.0001) (Table 4). The latter two were also identified as significant predictors of late mortality on stepwise multivariate analysis; COPD had a marginally significant effect (p<0.077), and reoperations, in fact, improved patient survival (Table 5).

Table 5-Stepwise Multiple Regression to Predict Late Deaths

Postoperative ischemic cardiomyopathy Postoperative stroke COPD Reoperation

Coefficient

Beta

F Ratio

p

SE

0.5501 0.2131 0.1259 -0.1977

0.4059 0.1606 0.1381 0.1672

30.401 4.756 3.083 4.837

0.000 0.029 0.077 0.028

0.0998 0.0977 0.717 0.0899

CHEST 1106 I 51 NOVEMBER, 1994

1351

CABG +CAROTID STENOSIS+ PVD

1 3 1 1212

2 3 4 5 6 7 8 9 10 11 12

100 90 80 en 70 a: 60 0 > 50 > 40 a: ::> 30 en 20 '#. 10 0

YEARS

Angina returned in 24 (14.2%) survivors (95% Cl: 0.5 to 5.7% / yr) by 3 months to 7 years after CABG. Most of the recurrences were within first 5 years and

>

-

Q7+1.5 114+3 114+3 6D+4.3 64+3.2

WITHOUT POST OP STROKE

-

WITH POST OP STROKE

66 +3

64 + 3 .2

53+ 5 .1

53+ 5.1

Q2+5.8 77~13

73+8 5Q+10 53+5.1

No.

NS p
.. •

,

,.

" 2 3 4 5 6 7 8 9 10 11 12 45

8

27 5

2

YEARS

FIGURE 4. Actuarial survival of the patients who developed postoperative major stroke vs remaining patients in the series.

were treated by angioplasty (n=20), graft angioplasty (n =7) , or reoperation (n =8). Freedom from all ischemic heart-related events (angina, infarction, CHF, cardiomyopathy, and deaths) at 12 years was 49%±3.4% (Fig 6) . The total incidence of significant strokes (residual weakness >6 wks) in this series up to 12 years .-:>-UNSTABlE AHGINA

7~

S YlAAS

TWO VESSEl ~A8f. 72"'o SYEM8 _._ L YEF < 310% G*A. 7 YlAR8

100 90 80 ~ 70 0 60 > > 50 a: 40 ~ 30 *" 20 10

~ OLOINFAACTIOH 80"'- STEAM

+-RECENT INFAACT10N et"A. t YEAAa -

L. M~ STEM t5~ 5 Y£N18

43%

1 3 1 1212

2 3 4 5 6 7 8 9 10 11 12

YEARS

FIGURE 5. Actuarial survival of the patients with unstable angina, preoperative myocardial infarction, left mainstem disease, and poor left ventricle (LVEF <30%).

. - WITHOUT POSTOPERATIVE ISCHEMIC CARDIOMYOPATHY+ CHF _

WITH POSTOPERATIVE ISCHEMIC CARDIOMYOPATHY+ CHF 74 +2.5

73 + 3 .2

p<0.01

YEARS 1112 3112 1 5 10

No.

-oQ6+1.6 Q3+3 Q2+2.5 75+4.5 73+3.2

29 + 7 .1

29 + 7 .1

. . .,. .. .. p<0.05 p<0.01

20+7.1

,., ,. , , 8 " 1 3 1 2 3 4 5 6 7 8 9 10 11 12 1212 151 24

138 2'37 22 21

,

72

YEARS

FIGURE 3. Actuarial survival of the subset of patients with ischemic cardiomyopathy vs remaining patients in series.

1352

1112 3112 1 5 10

-

0

Postoperative Events

100 90 80 en a: 70 0 60 > 50 a: 40 ::> en 30 '#. 20 10 0

YEARS

1 3 1 1212

FIGURE 2. Actuarial survival of all patients included in the series.

Actuarial survival for the entire series at 5 years was 72% (95% confidence interval [CI]: 66 to 79%) and at 12 years was 65% (95% Cl: 60 to 70%) (Fig 2). At 12 years, actuarial survival of the subsets of patients with preoperative CHF, COPD, or PVD postoperative MI and postoperative ischemic cardiomyopathy were 46%,49%,22%,37%, and 29% (Fig .3), respectively, being significantly lower than the remaining patients without these risk factors . Patients with postoperative stroke had a greater probability of deaths within first 12 months, particularly in patients with bilateral CS (Fig 4). At 7 years, actuarial survival in the patients with poor LV function (LVEF less than 30%) was 43% (Fig 5) . Long-term survival of the patients who received IMA graft or underwent reoperation for recurrent CAD was apparently improved. These procedures did not increase postoperative morbidity. Unstable angina, preoperative MI, and left mainstem disease had no detrimental effect on patient survival (Fig 5).

97 + 1 5

100 90 0 80 ~(/) w,_ 70 Q~ - ... 60 -'o 50 ~~ :E .. 40 oiil a: a: 30 ........ u.oa: 20 ...... a:u.o LL:J: 10 "' 0

o/o FREE FROM ALL CARDIAC EVENTS YEARS POSTOP 1 5 10 No.

175

1 3 1

1212

148

49

+ 3 .4

-oDh1.7 74+5 4Q*3.4

121

100

.

70

67

..

••

21

2 3 4 5 6 7 8 9 10 11 12

YEARS

FIGURE 6. Freedom from all ischemic heart-related events. Surgical Management in Coronary and Cerebrovascular Disease (Kaul eta/)

100 90 ~ (/) 80 a: 70 0 ~ 60 :::!! 50 ::;; 0 40 a: u. 30 :;; 0 20 0 UJ 10 UJ If 0

Table 6-Relationship Between Preoperative Neurologic Status and Postoperative and Late Neurologic Events*

TIA Preoperative neurologic status Asymptomatic (n=ll1 ) A Bilateral disease B Unilateral disease Lesion 80-85% TIA (n=49) Bilateral disease

Unilateral disease Lesion ~80% Major stroke (n = 15) Bilateral disease

Unilateral disease ~80 %

c

A B

c

A B

c

UJ

:.:: 0

Major Stroke After Surgery

n

17 12 39

3 Nil 4

lf

rf1.

lt+3 4

43

3

5

8 1 17

Nil 2

3 Nil 1

23

1

4

5 5

Nil Nil Nil

Nil Nil

4

Nil

2

UNILATERAL CAROTlD STENOSIS ~ ........ BILATERAL CAROTID STENOSIS 14.,._

BILATERAL CAE 12"'.

-

-o-

,. " , ,. .... "".

~~~¢:::~Lc~:~~~ :~~~~~

.

,. " " " " " " 1 3 1 2 3 4 5 6 7 1212 30

""

10S

69 63

53

97

73

M

38 43

30

31

38

::

}

p<

o.o7

.

, """ "'

8 9 10 11 12

YEARS

FIGURE 7. Actuarial representation of freedom from major stroke in subsets of patients with internal carotid disease.

tralateral stenosis <85%) (p<0.07) (Fig 7) . Fourteen in the latter subset of patients underwent prophylactic CAE for the progression of their contralateral disease at a mean interval of 4.0 ± 0.6 years after the initial CAE, which reduced the incidence of major stroke for these patients (Fig 8). Thirteen (7.4%) patients developed restenosis (11 with bilateral disease, 2 with unilateral disease) and underwent another CAE at 6.7 ± 1.1 years after the initial CAE. By multivariate analysis, systemic hypertension has the most important association with CAD and cerebrovascular disease (p<0.0001). The most important predictor of PVD in multivariate analysis was cigarette smoking (coefficient=0.2729, multisystem vascular disease, SE ± 0.061) (p<0.0001). Six patients in this subset of patients underwent balloon angioplasty for PVD within 6 weeks after combined CAE and CABG. Long-term survival of this subset of patients was least favorable (22% at 12 years) (Fig 9). At 12 years, freedom from symptomatic PVD was 7 4± 11 .3% . During this long follow-up, 36 more patients required surgical intervention for progression of their preexisting PVD (angioplasty 16, iliofemoral bypass 4, femoropopliteal bypass 12, and amputations 4) at 6 ± 1.8 years after the initial CABG and CAE.

*Major stroke=residual deficit persisting for more than 6 wks; A=bilateral internal carotid lesions ~80%; B=ipsilateral 80% contralateral occlusion (100%); C=ipsilateral ~80%, contralateral ~50% .

t Early stroke.

following surgery was 14.2% (n=25; 95% CI: 0.6 to 5.8%/yr). Although our symptomatic patients had a higher incidence of neurologic complications (Table 6), no independent predictor of major stroke was identified after CAE on the stepwise multivariate (Table 7) . The incidence of major stroke in the first postoperative month was higher in patient with bilateral disease than in patients with unilateral disease (5.7% vs 1.4%: NS). In patients with bilateral carotid disease, freedom from major postoperative stroke at 12 years was 73% when bilateral CAE was required (bilateral disease >85%) as compared with 88% when only unilateral CAE was required (con-

Table 1-Stepwise Multiple Regression to Predict Postoperative Stroke

TIA Age Sex, male Hypertension Diabetes Poor LV function CHF Unilateral carotid Bilateral CS Preoperative stroke

Coefficient (b)

Beta

F Ratio

p

SE

0.0637 0.0007 0.0763 0.0700 0.0637 0.0218 0.0807 0.0662 0.0641 0.0977

0.0818 0.0203 0.1001 0.953 0.0824 0.0200 0.0740 0.0943 0.0914 0.0748

0.981 0.041 1.418 1.266 0.963 0.053 0.768 0.025 0.024 0.802

0.675 0.820 0.234 0.261 0.670 0.803 0.614 0.847 0.851 0.625

0.0643 0.0033 0.0641 0.0622 0.0649 0.0945 0.0920 0.4150 0.4162 0.1091

CHEST / 106 / 5 / NOVEMBER, 1994

1353

HAZARD RATE X 10,000 (70% CL)

0.05 ~ '/. ··' ;:;

- - - BILATERAL CAE

~

BILATERAL DISEASE

...... UNILATERAL CAE

0.04

100 90

;,;..

>

-

'#. 20 10 0

~ ~

0.01

~

~

YEARS 1/12 3112 1 5 10

1

2

5

4

3

6

7

CAD+ CAROTID ST£NOSIS + PVD

8

- --

100 85+3.8 Q2+2.7 7h2 70+1 .7

85~2

Q2+2.7 02+3 5e+8.5 22+15

p<0.01

NS NS NS NS

22 + 15

p<0.01

No. 33 121

1 3 1212

1 12

CAD+ CAROTID ST£NOSIS

-

70+1.7

~ 30

0.02

~

....:

-

80

cr; 70 0 60 > 50 cr; 40

0.03

;.;.;:

95+3.6

(I)

UNILATERAL DISEASE

_,... IPSILATERAL CAE

100

1

28 •

21 78

,, 70

15

13

10

&

2

62

54

..

34

18

2 3 4 5 6 7 8 9 10 11 12 YEARS POST OP

9. Actuarial survival of the subset of patients with significant PVD vs remaining patients in the series. FIG URE

YEARS POST OP 8. Cumulative incidence of stroke in subsets of patients with internal carotid stenosis. Patients requiring bilateral carotid endarterectomy had highest incidence of major postoperative strokes. FIGURE

DISCUSSION

Incidence of Significant Coexistent CAD and Carotid Lesions A high incidence of symptomatic CAD (17 to 54%) has been reported in patients with internal CS, 16·17 but conversely on noninvasive Doppler scanning, only 6 to 16% patients with CAD were found to have a significant internal CS. 7·21 Noninvasive assessment of carotid lesions with oculopathysmography carotid

compression and Doppler examination has been used by most of the authors in the past and has been considered useful and fairly accurate.l 4 However, advantages of angiographic confirmation of vascular lesions before surgical intervention are obvious.ll We have used Doppler examination for general screening but angiography for confirmation in all patients. Before surgery, no serious sequelae or complications were encountered from routine angiography. Coexistence of significant CAD and cerebrovascular disease is not as common as has been projected by noninvasive assessment. Only 3.1% of our patients undergoing CABG in the last 12 years had significant internal CS, which is within the range (0.7 to 3.4%)

Table 8-Combined CAE & CABG First Author, yr

n

%Early Mortality

%Early Stroke

Author Recommendation

Bernhard,3 1972 Mehigan, 7 1977 Morris,5 1977 Enn ix, 16 1979 Rice,13 1980 Barnes, 18 1981 A collective review Reichart, 28 1982 Lawrie, 27 1982 Craver, 21 1982 Schwartz, 23 1982 CAE + CABG on bypass O'Donnell, 17 1983 Hertzer,ll 1983 Jones, 10 1984 Berkof£, 8 1984 Babu, 29 1985 Ivy,26 1986 Rosenthal, 12 1984 Matar, 2 1986 Newman, 25 1988 Cosgrove, 30 1986 Vermeulen,14 1992

16 29 44 51 54 1,161

Nil 13.8 Nil 5.1 Nil 4.7

Nil Nil 2.5 Nil 1.9 2.9

Yes Yes for high risk Yes Yes Yes Not for asymptomatic internal CS

15 51 68 73

Nil 3 Nil 10

Nil Nil 1.5 15

Yes Yes Yes Yes

22 331 132 16 62 5 24 32 12 38 230

4.5 5.7 3 Nil 4.8 Nil Nil Nil Nil 5.8 3.5

4.5 9* 1.6 4.5 1.6 20* Nil Nil Nil 5.9 8.7*

Selective approach Yes Selective approach No Yes No Yes Yes Not for asymptomatic internal CS No Yes

*A higher incidence of perioperative strokes was related to severe and bilateral disease. CAE & CABG:Carotid endarterectomy followed by coronary artery bypass grafting under same anesthesia.

1354

Surgical Management in Coronary and Cerebrovascular Disease (Kau/ eta/)

Table 9-Staged Approach for Coexistent Coronary and CS*

First Author, yr Bernhard, 3 1972 Mehigan, 7 1977 Morris,5 1977 Hertzer,24 1978 Ennix,16 1979 Barnes, 18 1981 A collective review O'Donnell,17 1983 Lawrie,27 1982 CAE first CABG first Rosenthal, 12 1984 Berkoff,8 1984 Newman,25 1988 A collective review Cosgrove,30 1986

%Early Mortality

% Early Stroke

15 23 35 59 77 322

20 Nil 20 1.7 18

6.6 4.3 Nil 1.7 5.2 3.1

No Yes for low-risk patients No Yes No Yes

17 77 84 22 5

Nil 18.2 3 4.5 Nil

Nil

Selective approach Suggested guidelines for staged and combined approach

9 20

No Yes Favors staged approach, especially for asymptomatic patients

26

2.8

11

Yes

n

Author Recommendation

*All authors listed in this table have used both staged and combined approach. Risk of myocardial infarction in CAE first approach has ranged between 4.6 and 26.7% and perioperative stroke rate was no less than patients undergoing combined CAE & CABG. reported in the recent literature.6·9•11 •13J4

Disease Pattern Patients included in this series obviously had advanced atherosclerosis. These patients are well known to have a higher incidence of left-main coronary disease (10 to 37%) , unstable angina (34 to 57%), triple-vessel CAD (40 to 80%) , impaired LV function (9 to 26%), CHF (7 to 10%), and a previous episode of MI (57 to 63%). 7·9-11 ·14·22·23 Our patients had a higher incidence of triple-vessel CAD (90%) and bilateral carotid vessel disease (60%) than has been reported earlier (34.5 to 50%).ll·14 Our patients with bilateral carotid disease had a male preponderance (76%), and 38% of these patients also had coexistent PVD.

Timing of Surgical Intervention To our knowledge, Bernhard et aP were the first to show a lower incidence of cardiac-related deaths and neurologic complications after a combined approach (CAE and CABG) , as compared with staged interventions. Since then, combined CAE and CABG have been successfully performed with a perioperative mortality ranging between 0% and 5.7%, and neurologic complications between 0% and 9% .2·4·7.10- 14 In patients with coexistent CAD and internal CS when CAE was performed first, postoperative MI developed in 10 to 38% 3·5·11 ,1 6 •24 and when only CABG was performed first, 0 to 20% developed postoperative stroke. 3·5·7·8·12 .1 6-18·24-28 Most published reports2-14·16·28 ·29 have preferred a combined surgical approach (CAE followed by CABG under the same anesthesia) for coexistent symptomatic and critical CAD and CS8.10,ll.l 8,25-27 over stage interventions (Tables 8 and 9). In a collective review of 20 articles,

Barnes and Marszalek 18 have shown that although the risk of MI was higher (11.8%) after staged intervention (CAE first), the incidence of postoperative stroke was no different after staged or combined procedures (3.1 vs 2.8: NS) . Based on these findings and in view of the fact that the cause of perioperative stroke is multifactorial, some of the authors have suggested that staged procedure (CABG first) is safe, especially in patients asymptomatic from carotid disease.8·10·18·25-28·30 The main criticism of these reports is that no attempt was made to correlate the location and the severity of the carotid lesion with the postoperative neurologic deficits. 31

Natural History of High-Grade CS and Prophylactic Value of CAE Recent studies have shown that in asymptomatic patients with high-grade CS, the stroke rate is 4 to 5% per year; in patients with TIA, 6% per year (10% in first year), and in patients with a history of stroke, it is 4 to 5% per year.32·33 Carotid endarterectomy provides a durable protection from ipsilateral stroke.31 After a successful carotid endarterectomy, stroke rate was reduced to 1.1 to 1.5% in asymptomatic patients, 1 to 2% in patients with TIA, and 3 to 4% in patients with established strokes.31 ·33 Asymptomatic high-grade CS should not be regarded as innocuous disease. 25 Late neurologic sequelae do occur in these patients without CAE. 31 ·34 ·35 In these patients, prophylactic CAE is justified, provided the risk of perioperative stroke is absolutely minimal in the hands of the surgical team and the risk without surgical intervention is clearly understood by the patient.l4,25,31 CHEST /1 06 I 5 I NOVEMBER, 1994

1355

Bilateral Carotid Stenosis Presence of bilateral CS, whether symptomatic or not, increases the risk of neurologic complications.l4·36-39 In patients with bilateral critical stenosis (>85%) and more so in patients with unilateral occlusion, the incidence of perioperative stroke has ranged between 15.6 and 20% .11,1 4,36-39 It has been suggested that most severe stenosis or the symptomatic side should be chosen first. 14 Bilateral CAE synchronous with CABG should be considered if both cerebral hemispheres are considered to be at risk.9.14 We have found cerebral angiography most useful in such cases. Bilateral carotid endarterectomy is generally not favored because of an element of risk due to intimal flap formation, postoperative occlusion, and minor airway or swallowing problems.21 It is possible to maintain adequate cerebral perfusion during cardiopulmonary bypass, 40 A1 but it is not advantageous to perform CAE during bypass.23 ·42

Perioperative Mortality and Morbidity As reported earlier,ll·14 early mortality of our patients undergoing combined CAE and CABG was marginally higher than our patients undergoing CABG along during the same period (3.4% vs 2.4%: NS) . In this series, early deaths were the result of MI occurring in patients with unstable angina and the result of preoperative stroke occurring in patients with bilateral carotid disease. Unlike previous reports,ll.l4·18A3left-main CAD and presence of L VEF >30% had no significant effect on the perioperative mortality of our patients. As reported earlier ,11·14 more late deaths occurred because of ischemic heart disease than strokes.l 4 In this series, other predictors of late mortality were COPD and PVD. All our patients with PVD also had triple-vessel CAD and 70% of them also had bilateral CS. This subset of our patients with multifocal atherosclerosis had least favorable long-term survival. As expected, 11 long-term survival of our patients with bilateral carotid disease was less favorable than our patients with unilateral disease. In this series, fatal MI occurred in 2.2% patients after combined CAE and CABG, which is within the range (0 to 5%) described earlier. 6·9·11 .1 3 Postoperative mortality and morbidity due to MI is generally lower after combined CAE and CABG than after staged intervention .25 In this series, incidence of perioperative stroke was four times higher in patients with bilateral carotid disease as compared with those who had unilateral carotid disease. In our patients with bilateral CS, an overall incidence of stroke was higher when lesions were critical on both sides than in patients who had unilateral critical stenosis. The latter underwent contralateral CAE when the lesion 1356

progressed. An elective contralateral CAE in this subset in patients has been shown to reduce the incidence of neurologic complications.31 ·43 As expected,ll,l4 our patients with a history of neurologic symptoms had a higher incidence of late strokes than our asymptomatic patients. Our study clearly shows that CAE reduces the incidence of neurologic deficits. In this series, recurrent CS after CAE developed in 7.4% as compared with 9 to 24% reported by others for a comparable study period. 43 Atherosclerosis is a progressive and multifocal disease. These patients should be carefully evaluated for recurrent symptoms, progression, and development of newer lesions and they should be treated accordingly. In our experience, combined CAE and CABG is the safest in technique, even in the presence of left-main CAD, recent MI, unstable angina, patients with poor LV function (LVEF, 30%), in those who required IMA as conduit, and in reoperation for recurrent CAD. Most recent reports 14·44 now favor combined over staged intervention. Most of our patients with coexistent PVD underwent peripheral vascular reconstruction as staged procedures. Synchronous approach is possible when symptoms dictate 15 Angioplasty was preferred for more proximal and less severe lesions and peripheral vascular reconstruction was preferred for more distal lesions. As expected, limb salvage rate was better for proximal disease than for distal disease. ACKNOWLEDGMENTS: The authors thank Melody Hamrick for secretarial help in preparation of this manuscript. REFERENCES

1 Gersh BJ, Rihal C, Roke TW, Ballard DJ. Evaluation and management of patients with both peripheral vascular and coronary artery disease. J Am Coli Cardiol 1991; 18:203-14 2 Matar AF. Concomitant coronary and cerebral revascularization under cardiopulmonary bypass. Ann Thorac Surg 1986; 41:431-5 3 Bernhard VM, Johnson WD, Peterson JJ. Carotid artery stenosis: association with surgery for coronary artery disease. Arch Surg 1972; 105:837-40 4 Urschel HC, Razzuk MA, Gardner MA. Management of concomitant occlusive disease of carotid and coronary arteries. J Thorac Cardiovasc Surg 1976; 72:829-34 5 Morris GC, Ennix CL Jr, Lawrie GM, Crawford ES, Howell JF. Management of coexistent carotid and coronary artery occlusive atherosclerosis. Cleve Clin Q 1977; 45:125-27 6 Aces JE, MacManus Q, Starr A. Myocardial revascularization and carotid endarterectomy: a combined approach. Ann Thorae Surg 1977; 23:560-63 7 Mehigan JT, Buch WS, Pipkin RD, Fogarty TJ. A planned approach to coexistent cerebrovascular disease in coronary artery bypass candidates. Arch Surg 1977; 112:1403-09 8 Berkoff HA, Turnip WD. Patient selection and results of simultaneous coronary and carotid artery procedures. Ann Thorac Surg 1984; 38:172-75 9 Schultz RD~ Sterpetti AV, Feldhaus RJ. Early and late results in patients with carotid disease undergoing myocardial revascularization. Ann Thorac Surg 1988; 45:603-09 10 Jones EL, Craver JM, Michalik RA, et al. Combined carotid and

Surgical Management in Coronary and Cerebrovascular Disease (Kaul eta/)

11

12 13 14

15

16

17 18 19 20 21 22

23 24 25

26 27

coronary operations: when are they necessary? J Thorac Cardiovasc Surg 1984; 87:7-16 Hertzer NR, Loop FD, Taylor PC, Beven EG. Combined myocardial revascularization and carotid endarterectomy: operative and late results in 331 patients. J Thorac Cardiovasc Surg 1983; 85:577-89 Rosenthal D, Caudil DR, Lamis PA, Logan W, Stanton PE. Carotid and coronary arterial disease: a rational approach. Am J Surg 1984; 50:233-35 Rice PL, Pifarre R, Sullivan HJ, Montoya A, Bakhos M. Experience with simultaneous myocardial revascularization and carotid endarterectomy. J Thorac Cardiovasc Surg 1980; 79:922-25 Vermeulen FEE, Hamerlijnck RPHM, Defauw JJAM, Ernest SMPG. Synchronous operation for ischemic cardiac and cerebrovascular disease: early results and long-term follow-up. Ann Thorac Surg 1992; 53:381-90 Carrel J, Niederhauser U, Pasic M, Gallino A, Von Segesser L, Turina M. Simultaneous revascularization for critical coronary and peripheral vascular ischemia. Ann Thorac Surg 1991; 52:805-91 Ennix CL, Lawrie GM, Morris GC, et al. Improved results of carotid endarterectomy in patients with symptomatic coronary artery disease: an analysis of 1,546 consecutive operations. Stroke 1979; 10:122-25 O'Donnell TF, Callow AD, Willet C, Payne D, Cleveland RJ. The impact of coronary artery disease on carotid endarterectomy. Ann Surg 1983; 198:705-12 Barnes R, Marszalek P. Asymptomatic carotid disease in cardiovascular surgical patients: is prophylactic endarterectomy necessary? Stroke 1981; 12:497-500 Emery RW, Cohn LH, Whittemore AD, Mannick JA, Couch NP, Collins JJ Jr. Coexistent carotid and coronary artery disease: surgical management. Arch Surg 1983; 118:1035-38 Mantel N. Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 1966; 50:163-70. Craver JM, Murphy DA, Jones EL, et al. Concomitant carotid and coronary reconstruction. Ann Surg 1982; 195:712-20 Vigna GB, Bolzan M, Romagnoni F, et al. Lipids and other risk factors selected by discriminant analysis in symptomatic patients with supra-aortic and peripheral atherosclerosis. Circulation 1992; 85:2205-11 Schwartz RL, Garrett JR, Karp RB, Kouchoukos NT. Simultaneous myocardial revascularization and carotid endarterectomy. Circulation 1982; 66(suppl1):97-101 Hertzer NR, Loop FD, Taylor PC, Beven EG. Staged and combined surgical approach to simultaneous carotid and coronary vascular disease. Surgery 1978 84:803-11 Newman DC, Hicks RG. Combined carotid and coronary artery surgery: a review of the literature. Ann Thorac Surg 1988; 45:574-81 Ivy TD. Combined carotid and coronary disease: a conservative strategy. J Vase Surg 1986; 3:687-89 Lawrie GM, Morris GC. Survival after coronary artery bypass surgery in specific patient groups. Circulation 1982; 65(suppl

2):43-8 28 Reichart B, Becker HM, Autenrieth G, et al. Stenosis of supraaortic branches combined with coronary artery disease: onestage surgical treatment. Thorac Cardiovasc Surg 1982; 30:269-72 29 Babu SC, Shah PM, Singh BM, Semel L, Clauss RH, Reed GE. Coexisting carotid stenosis in patients undergoing cardiac surgery: indications and guidelines for simultaneous operations. Am J Surg 1985; 150:207-11 30 Cosgrove DM, Hertzer NR, Loop FD. Surgical management of synchronous carotid and coronary artery disease. J Vase Surg 1986; 3:690-94 31 Hertzer NR, Arison R. Cumulative stroke and survival10 years after carotid endarterectomy. J Vase Surg 1985; 2:661-68 32 Henneerici M, Hulsbomer HB, Hefter H, Lummerts D, Rautenberg W. Natural history of asymptomatic extracranial arterial disease: results of a long-term prospective study. Brain 1987; 110:779-91 33 Quinones-Baldrich WJ, Moor WS. Results of medical and surgical therapy for extracranial arterial occlusive disease. In: Rutherford RB, ed. Vascular Surgery. 3rd ed (vol2). Rutherford RB, ed. Philadelphia: WB Saunders, 1989; 1472-78 34 Lees CS, Hertzer NR. Postoperative stroke and late neurological complications after carotid endarterectomy. Arch Surg 1981; 116:1561-68 35 Cooperman M, Martin EW, Evans WE. Significance of asymptomatic carotid bruits. Arch Surg 1978; 113:1339-40 36 Thompson JE, Austin DJ, Patman RD. Carotid endarterectomy for cerebrovascular insufficiency: long-term results in 592 patients followed up to 13 years. Ann Surg 1970; 172:663-79 37 Brenner BJ, Brief DK, Alpert J, et al. A 4-year experience with preoperative noninvasive carotid evaluation of 2026 patients undergoing cardiac surgery. J Vase Surg 1984; 1:326-38 38 MacKay WC, O'Donnell TF Jr, Callow AD. Carotid endarterectomy contralateral to an occluded carotid artery: preoperative risk and late results. J Vase Surg 1990; 11:778-85 39 Hammacher ER, Eikoom BC, Bast TJ, DeGeest R, Vermeulen FEE. Surgical treatment of patients with carotid artery occlusion and a contralateral stenosis. J Cardiovasc Surg 1984; 25:513-17 40 Allen B. Cardiopulmonary perfusion and cerebral blood flow in bilateral carotid disease. Ann Thorac Surg 1991; 51:532-33 41 Johnson P, Algotsson L, Ryding E, Stahl E, Messeter K. Cardiopulmonary perfusion and cerebral blood flow in bilateral carotid artery disease. Ann Thorac 1991; 51:579-84 42 Weiss SJ, Sutter FP, Sannon TO, Goldman SM. Combined cardiac operation and carotid endarterectomy during aortic cross clamping. Ann Thorac Surg 1992; 53:813-16 43 Gardner TJ, Horneffer PJ, Manolio TA, Hoff SJ, Pearson TA. Major stroke after coronary artery bypass surgery: changing magnitude of the problem. J Vase Surg 1986; 3:684-87 44 Kirklin JW, Barratt-Boyces BG. Stenotic arteriosclerotic coronary artery disease. In: Kirklin JW, Barratt-Boyces BG, eds. Cardiac surgery. 2nd ed (vol1). New York: Churchill Livingstone, 1993; 285-342

CHEST /106/5/ NOVEMBER, 1994

1357