Gen Thorac Cardiovasc Surg DOI 10.1007/s11748-014-0371-4
ORIGINAL ARTICLE
Bilateral internal thoracic artery grafting for peripheral arterial disease patients Taro Nakatsu • Nobushige Tamura • Shigeki Yanagi • Shoichi Kyo • Takaaki Koshiji Ryuzo Sakata
•
Received: 18 November 2013 / Accepted: 6 January 2014 Ó The Japanese Association for Thoracic Surgery 2014
Abstract Background Peripheral arterial disease (PAD) is known to have a poor prognosis. Some reports have also described poor late results after coronary artery bypass grafting (CABG). However, there have been few reports about whether bilateral internal thoracic artery grafting improves the long-term survival of PAD patients after CABG. We performed this study to clarify whether or not this is the case. Methods One hundred and thirty-six PAD patients who underwent elective CABG with two or more grafts were enrolled in this study. Patients were divided into two groups, 71 patients in the bilateral internal thoracic artery (BITA) group and 65 patients in the single internal thoracic artery (SITA) group. The maximum follow-up period was 19 years, with a mean of 5.7 ± 4.4 years. Results We investigated the long-term results based on three factors; survival, freedom from cardiac death, and freedom from cardiac events. The 3-, 5- and 10-year survival rates in the BITA group were 83.0, 74.2, and 43.1 %, respectively. And those in the SITA group were 79.4, 67.7, and 32.3 %, respectively. There were no significant T. Nakatsu (&) N. Tamura S. Yanagi S. Kyo Department of Cardiovascular Surgery, Kumamoto Central Hospital, 1-5-1 Tainoshima, Kumamoto 862-0965, Japan e-mail:
[email protected] T. Koshiji Second Department of Surgery, Faculty of Medical Sciences, University of Fukui, 23-3 Matsuokashimoaizuki, Eiheiji-cho, Yoshida-gun, Fukui Prefecture 910-1193, Japan R. Sakata Department of Cardiovascular Surgery, Graduate School of Medicine, Kyoto University, 54 Kawaharacho, Shogoin, Sakyo-ku, Kyoto 606-8507, Japan
differences between the two groups (p = 0.5843). There were also no statistically significant differences between the two groups in terms of the freedom from cardiac death (p = 0.8589) or in the freedom from cardiac events (p = 0.9445). Conclusion No long-term advantage was observed for CABG with BITA in comparison to SITA alone in patients with PAD. Keywords CABG, arterial grafts Peripheral vascular disease (all non-thoracic arteries, except carotid)
Introduction Coronary artery bypass grafting (CABG) is the most common cardiac surgery, and it is performed on from 18,000 to 20,000 cases per year in Japan [1]. Peripheral arterial disease (PAD) is the end stage of arteriosclerosis, and it is known to have a poor prognosis [2]. Several reports have emphasized that the surgical mortality associated with coronary artery bypass grafting for patients with PAD is higher than that of patients without PAD [3, 4]. PAD patients have a worse prognosis in not only the early results, but also in the long-term survival, than patients without PAD [5, 6]. Generally, CABG with the bilateral internal thoracic artery (BITA) provides a better long-term result after the surgery than CABG with a single internal thoracic artery (SITA) [7–10]. However, it is unclear whether CABG using BITA improves the longterm survival of PAD patients or not. We previously reported that patients requiring hemodialysis, who have a poor prognosis similar to PAD patients, had no long-term advantages associated with BITA grafting compared with SITA grafting [11]. Concerning the graft strategy for PAD
123
Gen Thorac Cardiovasc Surg
patients, it is important to know whether BITA would lead to a better prognosis than SITA when performing CABG. Therefore, we investigated the long-term results after CABG for PAD patients, and compared the long-term survival in patients after BITA grafting and SITA grafting.
Patients and methods The study was approved by our local institutional research ethics committee, and individual patient’s consent was waived. Data were obtained from the database of the Department of Cardiovascular Surgery of Kumamoto Central Hospital, in which the data were registered prospectively. Between January 1990 and June 2010, 2718 patients underwent CABG at the Kumamoto Central Hospital, and of these patients, 199 had PAD. We included the 136 patients who underwent elective CABG with two or more anastomoses using a single or bilateral internal thoracic artery. Seventy-one patients underwent BITA grafting, and sixty-five patients had a SITA graft. Patients who had concomitant lower extremity procedures were included in this study. This series also included four re-operation cases. Emergency cases, CABG with only a single anastomosis, and cases with other cardiac procedures were excluded from this study. Definitions After 2002, we preoperatively checked the Ankle-brachial systolic pressure index (ABI) as a screening method for PAD. We defined PAD as an ABI \0.90 based on the definition of PAD in the TASCII guidelines [12]. We also diagnosed PAD in patients with a history of lower limb revascularization or amputation. Before 2002, PAD was diagnosed based on the stenosis of the abdominal aorta or arteries below the aortic bifurcation by angiography or echocardiography. Surgical technique After a median sternotomy, the ascending aorta was evaluated with an epiaortic echocardiogram to assess the patients for the presence of atherosclerosis or calcification to decide whether or not to clamp the aorta. Internal thoracic arteries were harvested in a pedicled manner before December 1999 and in a skeletonized manner thereafter. Prior to 2000, if the patient’s aorta could not be clamped due to severe atherosclerosis or calcification, the operation was performed under ventricular fibrillation. After 2000, these patients received either on-pump beating CABG or off-pump CABG. Eighty-two patients were surgically treated under aortic clamping. Eleven patients underwent
123
the CABG procedure under ventricular fibrillation. Onpump beating CABG was performed in nine patients. Offpump CABG was undertaken in thirty-four patients. Concomitant lower extremity procedures were performed in thirteen patients, seven patients in the BITA group and six patients in the SITA group. A femoro-femoral bypass was undertaken in six patients, and a femoropopliteal bypass was performed in seven patients. Follow-up All patients were followed up based on their records and telephone interviews. One hundred percent of patients could be followed up. The mean follow-up duration was 5.7 ± 4.4 years, with a range from 2 months to 19.3 years. The survival, cardiac death and cardiac event rates were compared between the two groups. Cardiac death in this study included in-hospital deaths, cardiogenic deaths, sudden deaths, and death from unknown causes. The cardiac events included cardiac death, myocardial infarction, recurrent angina, repeated CABG, repeated percutaneous coronary intervention, and congestive heart failure. Statistical analysis Categorical variables were described as numbers and percentages and they were compared using the v2 exact test. Continuous variables were expressed as the mean ± standard deviations and analyzed with 2-tailed Student’s t tests. The survival, freedom from cardiac death, and freedom from cardiac events were estimated by the Kaplan–Meier method and compared with log-rank statistics. A univariate and multivariate Cox proportional hazard analysis was used to examine the influence of preoperative or surgeryrelated factors on survival. If the p value was \0.10 in the univariate Cox proportional hazard analysis, the factor was included in the multivariate Cox proportional hazard analysis. A value of p \ 0.05 was considered to be statistically significant. All statistical analyses were performed using the StatView software program, version 5.0 (SAS Institute, Cary, NC).
Results Table 1 shows the characteristics of the two groups. Since better long-term survival was expected, we tended to use BITA grafts for younger patients. Therefore, the BITA group was significantly younger than the SITA group. The length of the operation in the BITA group was longer than that of the SITA group, because of the time required to harvest the BITA. Otherwise, there were no significant differences between the two groups with regard to the
Gen Thorac Cardiovasc Surg Table 1 Patients’ characteristics
Table 2 Graft designs
BITA n = 71
SITA n = 65
P value
LITA
RITA
Male
60 (85 %)
51 (78 %)
Ns
LAD
46a
65 %
23
Age
68 ± 7.7
71 ± 6.5
0.0085
Diag or HL
6
8%
15
Hypertension
51 (72 %)
52 (80 %)
Ns
LCx
19
27 %
32
Diabetes mellitus
44 (62 %)
38 (58 %)
Ns
RCA
0
0
1
1%
Insulin
15 (21 %)
14 (22 %)
Ns
Hyperlipidemia
38 (54 %)
24 (37 %)
Ns
LAD
49d
75 %
7
11 %
Cerebrovascular disease Cre
20 (28 %) 2.4 ± 2.7
14 (22 %) 2.3 ± 2.8
Ns Ns
Diag or HL
4
6%
0
0
LCx
5
8%
0
0
Hemodialysis
11 (16 %)
13 (20 %)
Ns
RCA
0
0
0
0
Obesity
6 (8.4 %)
7 (11 %)
Ns
Redo
0
3 (4.6 %)
Ns
LAD left anterior descending artery, Diag diagonal branch, HL high lateral branch, LCx left circumflex artery, RCA right coronary artery
BITA
b
33 % 21 % c
45 %
SITA
Preoperative ABI
0.69 ± 0.14
0.73 ± 0.18
Ns
a
History of amputation
2 (2.8 %)
1 (1.5 %)
Ns
b
Including 2 composite grafts and 3 sequential anastomosis cases
Preoperative revascularization of lower extremities Concomitant lower extremities procedure
6 (8.5 %)
7 (10.8 %)
Ns
c
Including 7 composite grafts and 1 sequential anastomosis case
d
Including 7 sequential anastomosis cases
7 (9.9 %)
6 (9.2 %)
Ns
Usage of gastroepiploic artery
20 (28 %)
15 (23 %)
Ns
Usage of radial artery
10 (14 %)
8 (12 %)
Ns
EF \40
6 (8.5 %)
4 (6.2 %)
Ns
EF 4–60
24 (34 %)
23 (35 %)
Ns
EF [60
39 (55 %)
33 (51 %)
Ns
Preoperative LVDd
51 ± 6.1
49 ± 7.0
Ns
LMT disease
31 (44 %)
22 (34 %)
Ns
Two vessels disease
17 (24 %)
18 (28 %)
Ns
Three vessel disease
50 (70 %)
45 (69 %)
Ns
Operative time CPB time
371 ± 99 134 ± 54
321 ± 60 123 ± 36
0.001 Ns
Clamp time
68 ± 46
70 ± 40
Ns
OPCAB
21 (30 %)
14 (22 %)
Ns
ONPBCAB
6 (8.5 %)
3 (4.6 %)
Ns
CABG Vf
6 (8.5 %)
5 (7.7 %)
Ns
Anastomoses
3.4 ± 1.0
3.2 ± 0.8
Ns
IABP
3 (4.2 %)
0
Ns
ABI ankle-brachial systolic pressure index, EF ejection fraction, LVDd left ventricular diameter of diastric phase, LMT left main trunk, CPB cardiopulmonary bypass, OPCAB off-pump coronary artery bypass, ONPBCAB on-pump beating coronary artery bypass, Vf ventricular fibrillation, IABP intra-aortic balloon pumping
patients’ characteristics, such as the presence of hypertension, diabetes mellitus or hyperlipidemia and the preoperative cardiac function, number of diseased vessels and methods of grafting. Preoperative revascularizations for lower extremities, such as percutaneous transluminal angioplasty (PTA) or surgical bypass grafting, were performed in six patients (8.4 %) from the BITA group and seven patients (10.8 %)
Including 6 sequential anastomosis cases
from the SITA group. Preoperative histories of amputation of lower extremities were found in two patients in the BITA group, and one patient in the SITA group. The mean ABI of the worse side for the BITA group was 0.69 ± 0.14 and that for the SITA group was 0.73 ± 0.18. There were no statistically significant differences between the two groups in terms of the preoperative severity of PAD. Graft design The total mean number of bypass grafts was 3.3 ± 0.9 and 3.4 ± 1.0 for the BITA group, and 3.2 ± 0.8 for the SITA group. There were no apparent differences between the two groups in terms of the number of grafts. Table 2 demonstrates the graft design. Sixty-five percent of the patients in the BITA group underwent left internal thoracic artery grafting to the left anterior descending branch, and right internal thoracic artery grafting to the left circumflex branch, diagonal branch or high lateral branch. The other 35 % of the patients in the BITA group underwent right internal thoracic artery grafting to the left anterior descending branch, and left internal thoracic artery grafting to the left circumflex branch, diagonal branch or high lateral branch. In the SITA group, 86 % of the patients underwent internal thoracic artery grafting to the left anterior descending branch. Early results Table 3 shows the surgical results. Two patients in the BITA group and two in the SITA group died in the hospital. In the BITA group, one patient died due to
123
Gen Thorac Cardiovasc Surg Survival (%)
Table 3 Operative results BITA
SITA
P value
Mortality
2 (2.8 %)
2 (3.1 %)
NS
Resternotomy
0
1 (1.5 %)
NS
Stroke
5 (7.0 %)
4 (6.2 %)
NS
Mediastinitis
3 (4.2 %)
2 (3.1 %)
NS
Leg complication
2 (2.8 %)
5 (7.7 %)
100 80 BITA SITA
60 p=0.5843
40
NS
20
Table 4 Graft patency BITA
0
(%)
Patent grafts
Graft number
LITA
61
62
RITA
62
62
100
GEA
17
17
100
RA SVG
98.4
SITA
(%)
p value
Patent grafts
Graft number
57
57
100
NS
3
3
100
–
12
12
100
–
100
7
7
100
8
8
53
53
100
92
95
96.8
0
7.5
10
12.5
15
17.5
BITA
69
2.5 49
5 38
21
13
5
1
1
SITA
63
47
32
15
8
8
4
1
20 year
Fig. 1 Survival. No statistical differences were observed regarding patient survival (p = 0.5843). Dotted line single internal thoracic artery (SITA), solid line bilateral internal thoracic artery (BITA)
– NS
mediastinitis and the other died of a perioperative myocardial infarction. In the SITA group, one patient died of multi-organ failure and the other patient died of aspiration pneumonia. Wound complications involving the legs occurred in seven patients, 5.1 % of the entire cohort. No differences were observed between the two groups with regard to wound complications of the legs. Nine strokes were observed in or after the surgery, which involved 6.6 % of all patients. There were no differences between the two groups in terms of the incidence of stroke. Sixty-two patients in the BITA group (87.3 %) and sixty patients in the SITA group (92.3 %) were examined for the patency of the grafts by coronary arteriography or computed tomography before they were discharged (Table 4). The graft patency of the BITA group for the left internal thoracic artery was 98.4 %, and was 100 % for the right internal thoracic artery, gastroepiploic artery, radial artery and saphenous vein grafts. The graft patency of the SITA group was 100 % for the left internal thoracic artery, right internal thoracic artery, gastroepiploic artery and radial artery, while that of the saphenous vein grafts was 96.8 %. There were no significant differences in the graft patency between the two groups (p = 0.3356 for the left internal thoracic artery and p = 0.1912 for the saphenous vein grafts). Late results About PAD There were thirteen patients who required PTA in the follow-up period in the BITA group, and four patients in
123
No. at risk
the SITA group, p = 0.0323. There were eight patients who underwent surgical revascularization of lower extremities during the follow-up period in the BITA group, and four patients who required surgical revascularization in the SITA group, p = 0.2936. A total of eight and five legs had to undergo leg amputation during the follow-up period due to PAD in the BITA and SITA groups, respectively (p = 0.479). Survival Figure 1 shows the survival curves. The survival analysis included the hospital mortality. The 3-, 5- and 10-year survival rates in the BITA group were 83.0 ± 4.7, 74.2 ± 5.6 and 43.1 ± 7.5 %, respectively. Those in the SITA group were 79.4 ± 6.2, 67.7 ± 6.2 and 32.3 ± 7.9 %, respectively. There were no significant differences in the survival between the two groups (p = 0.5843 in the log-rank test). Although the younger cohort generally tends to have longer survival, no survival advantage was observed in the BITA group which was younger than the other group. There were 66 patients who died during the follow-up period. Fourteen patients died of an infection, which was the most common cause of death. Thirteen patients died of cardiac causes, nine patients died of cerebral events, such as cerebral infarction or cerebral hemorrhages, and five patients died of malignancies. Seven patients died suddenly, and then the accurate causes of death were unknown. There was no information available about the causes of death for another 12 patients. In the univariate Cox proportional hazard model examining the long-term survival, the patient age, chronic
Gen Thorac Cardiovasc Surg Table 5 Cox proportional hazard model for survival Univariate analysis
Multivariate analysis
HR
95 % CI
P
BITA
0.872
0.537–1.418
NS
Gastroepiploic artery
0.97
0.557–1.687
NS
Age
1.049
1.007–1.093
0.0222
Female
1.061
0.567–1.987
NS
Hypertension
1.531
0.826–2.837
NS
Diabetes mellitus
1.484
0.893–2.466
NS
Insulin Hyperlipidemia
1.536 0.796
0.87–2.717 0.479–1.325
NS NS
Cerebrovascular disease
1.157
0.645–2.077
NS
Hemodialysis
2.327
1.343–4.032
0.0026
LVDd
0.97
0.911–1.034
NS
EF \ 40
1.701
0.726–3.989
NS
EF 40-60
1.373
0.824–2.288
NS
EF [ 60
0.759
0.467–1.234
NS
Obesity
1.035
0.445–2.406
NS
Ope time
1.002
0.998–1.005
NS
CPB time
0.999
0.992–1.007
NS
Clamp time
0.995
0.988–1.002
NS
ABI
1.014
0.067–15.267
NS
Concomitant lower extremity procedure
0.935
0.44–1.984
NS
Radial artery
0.677
0.271–1.692
NS
Anastomoses LMT disease
0.873 1.257
0.646–1.181 0.77–2.049
NS NS
IABP
4.536
1.067–19.278
0.0405
Redo
2.281
0.54–9.64
NS
Stroke
3.303
1.565–6.97
0.0017
Gastrointestinal disorders
1.776
0.553–5.704
NS
Leg complication
0.998
0.362–2.755
NS
HR
95 % CI
P
1.068
1.018–1.12
0.007
3.503
1.906–6.439
\0.0001
1.596
0.353–7.215
NS
2.783
1.143–6.778
0.0242
LVDd left ventricular diameter of diastric phase, EF ejection fraction, ABI ankle-brachial systolic pressure index, LMT left main trunk, IABP intra-aortic balloon pumping
hemodialysis, insertion of an intra-aortic balloon pump and stroke in or after the surgery all had values of p \ 0.1. The multivariate analysis revealed that age, chronic hemodialysis and perioperative stroke were the predictors of longterm death (Table 5). Freedom from cardiac death Figure 2 shows the freedom from cardiac death curves. The 3-, 5- and 10-year rates of freedom from cardiac death in the BITA group were 92.2 ± 3.4, 86.1 ± 4.6 and 63.4 ± 8.5 %, respectively. Those of the SITA group were 93.7 ± 3.1, 86.4 ± 4.9 and 60.6 ± 10.0 %, respectively. No significant differences between the two groups were observed (p = 0.8589 in the log-rank test). In the univariate Cox proportional hazard model, six factors, including
high age, chronic hemodialysis, low ejection fraction (\40 %), use of intra-aortic balloon pumping, and stroke during or after the surgery had values of p \ 0.1. After analyzing these six factors by the multivariate study, chronic hemodialysis, a low ejection fraction (\40 %), and stroke were revealed to be predictors of long-term cardiac deaths. Freedom from cardiac events Figure 3 demonstrates the freedom from cardiac event curves. The rates of freedom from cardiac events in the BITA group at 3, 5 and 10 years after the procedure were 81.7 ± 4.8, 66.6 ± 6.5 and 41.3 ± 8.6 %, respectively. Those of the SITA group were 75.2 ± 5.6, 67.1 ± 6.7 and 41.7 ± 10.2 %, respectively. No significant differences
123
Gen Thorac Cardiovasc Surg Freedom from cardiac death(%)
100 80
BITA SITA
p=0.8589
60 40 20 0
0
No. at risk
2.5
5
7.5
10
12.5
15
17.5
BITA
69
49
38
21
13
5
1
1
SITA
63
47
32
15
8
8
4
1
20 year
Fig. 2 Freedom from cardiac death (including in-hospital deaths, cardiogenic deaths, sudden deaths, and death from unknown causes). No statistical differences were observed between the two groups (p = 0.8589). Dotted line single internal thoracic artery (SITA), solid line bilateral internal thoracic artery (BITA)
Freedom from cardiac event(%)
100
80 BITA SITA
60
p=0.9445
40 20
0 No. at risk
0
2
4
6
8
10
12
14
BITA
68
47
37
18
12
8
3
3
16 0
SITA
63
41
27
16
9
6
4
1
0
18 year
Fig. 3 Freedom from cardiac events (including cardiac death, myocardial infarction, recurrent angina, repeated CABG, repeated percutaneous coronary intervention, and congestive heart failure). No statistical differences were observed between the two groups (p = 0.9445). Dotted line single internal thoracic artery (SITA), solid line bilateral internal thoracic artery (BITA)
were found between the two groups (p = 0.9445 by the log-rank test).
Comments We demonstrated that the patients with PAD, who have a poorer prognosis than general patients, did not derive any advantage in terms of the long-term outcomes if they underwent BITA grafting compared to SITA grafting. There were some reports that emphasized even the elderly
123
patients might have better results if the arterial grafts were used [13]. On the other hand, we previously reported a similar tendency to this study in patients with end-stage renal disease [11]. The advantage of BITA grafting for long-term survival is reported to appear in the second decade [14]. The poor results for the 10-year survival rate among PAD patients after CABG may be the reason for the small differences between the two groups obtained in this study. Patients with PAD have previously been reported to have worse long-term outcomes compared with those without PAD [5, 6]. However, while we knew that PAD patients generally have a poor outcome after CABG, it was not known whether CABG with BITA could improve the long-term outcomes [3, 4]. In this study, we found that BITA provided no advantage over SITA for CABG with regard to the long-term survival in PAD patients. PAD patients had previously been reported to have higher mortality rates even after excluding the impact of cardiovascular diseases [1]. In this study, 66 patients had died during the follow-up period. Approximately a half of the patients died of non-cardiovascular causes, excluding 13 cardiac death patients, seven sudden death patients and 12 patients who died of unknown causes. BITA grafting may not reduce the mortality of these patients. Therefore, these non-cardiac deaths might diminish the long-term advantage of BITA grafting. BITA is unnecessary unless no other grafts can be harvested. PAD patients have been reported to have poorer outcomes than those without PAD, so it would be better to perform a simpler operation with a shorter duration. However, the patients who underwent BITA grafting did not have more early complications than patients with SITA grafting, so there are no contraindications for performing BITA grafting, especially in cases where there are no other grafts available to harvest. Hemodialysis was found to be one of the predictors of late death. There were 24 hemodialysis patients in this study. To investigate the hypothesis that a worse prognosis of these hemodialysis patients diminished the advantages of BITA grafting, as we previously reported, we performed a survival analysis among the patients without hemodialysis. As a result, no survival advantage was found in 60 BITA grafting patients in comparison to 52 SITA grafting patients (p = 0.8079 at log-rank test). In this study, we found that the mean 10-year survival rate of the BITA group was 43.1 %, and that of the SITA group was 32.3 % after CABG. These results were worse than those of previous studies. For example, Chu and his colleague reported a 72.9 ± 4.1 % 9-year survival rate after CABG among PAD patients, and van Straten and his colleague mentioned that the 9-year survival of PAD patients after CABG was 51.4 ± 3.5 % [5, 6]. The cause of the poor prognosis in this study may have been the more severe background and the higher mean age of our cohort.
Gen Thorac Cardiovasc Surg
For example, 24 out of the 136 patients were dependent on chronic hemodialysis, and 82 patients (60 %) had diabetes mellitus. In addition, the mean age of our cohort was 69.4 ± 7.32, which was higher than 63.2 ± 8.6 % [5] or 67.3 ± 8.2 % [6]. In this analysis, there were nine strokes that occurred during or after the surgery, which comprised 6.6 % of the whole cohort. PAD patients have a higher risk for stroke after CABG than general patients after CABG, whose rate of stroke is about 1–4 % [15–17]. We do not have any preoperative echo data investigating the carotid arteries, but there were 34 patients who had a previous stroke history before CABG. This frequency is high, since it is approximately 25 % of all the patients. PAD patients often have progressive systemic arteriosclerosis and the stenosis of their carotid and intracranial arteries might have progressed, and atheroma of their ascending aortas might lead to an embolism. These factors likely contributed to the high frequency of stroke after the surgery among the PAD patients. Nine patients died of cerebral events during the follow-up period in this study. This may be supported by the findings of Banerjee and colleague indicating that PAD patients had a higher risk of stroke than the normal population [18]. The major limitation of this study is that it was a single institutional retrospective observational study. The small sample size is also a limitation. In conclusion, for PAD patients, BITA grafting does not contribute to any improvement in the long-term survival, avoidance of cardiac death or avoidance of cardiac events in comparison to SITA grafting. Conflict of interest interest exists.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
The authors have declared that no conflict of
References 1. Sakata R, Kuwano H, Yokomise H. Hospital volume and outcomes of cardiothoracic surgery in Japan: 2005–2009 national survey. Gen Thorac Cardiovasc Surg. 2012;60:625–38. 2. Pande RL, Perlstein TS, Beckman JA, Creager MA. Secondary prevention and mortality in peripheral artery disease: National Health and Nutrition Examination Study, 1999–2004. Circulation. 2011;124:17–23. 3. Higggins TL, Estafanous FG, Loop FD, Beck GJ, Blum JM, Paranandi L. Stratification of morbidity and mortality outcome by
15.
16.
17.
18.
preoperative risk factors in coronary artery bypass patients. A clinical severity score. JAMA. 1992;267:2344–8. O’Connor GT, Plume SK, Olmsted EM, et al. Multivariate prediction of in-hospital mortality associated with coronary artery bypass graft surgery. Northern New England Cardiovascular Disease Study Group. Circulation. 1992;85:2210–8. Chu D, Bakaeen FG, Wang XL, Dao TK, LeMaire SA, Coselli JS, Huh J. The impact of peripheral vascular disease on long-term survival after coronary artery bypass graft surgery. Ann Thorac Surg. 2008;86:1175–80. van Straten AH, Firanescu C, Soliman Hamad MA, et al. Peripheral vascular disease as a predictor of survival after coronary artery bypass grafting: comparison with a matched general population. Ann Thorac Surg. 2010;89:414–20. Lytle BW, Blackstone EH, Loop FD, et al. Two internal thoracic artery grafts are better than one. J Thorac Cardiovasc Surg. 1999;117:855–72. Ura M, Sakata R, Nakayama Y, Arai Y, Saito T. Long-term results of bilateral internal thoracic artery grafting. Ann Thorac Surg. 2000;70:1991–6. Pevni D, Uretzky G, Mohr A, et al. Routine use of bilateral skeletonized internal thoracic artery grafting: long-term results. Circulation. 2008;118:705–12. Taggart DP, D’Amico R, Altman DG. Effect of arterial revascularization on survival: a systematic review of studies comparing bilateral and single internal mammary arteries. Lancet. 2001;358:870–5. Nakatsu T, Tamura N, Sakakibara Y, Hagio K, Ishigami M. Long-term survival after coronary arterial grafts in patients with end-stage renal disease. Ann Thorac Surg. 2010;90:738–43. Norgren L, Hiatt WR, Dormandy JA, Nehler MR, Harris KA, Fowkes FG, TASC II Working Group. Inter-society consensus for the management of peripheral arterial disease (TASCII). J Vasc Surg. 2007;45(Suppl S):S5–67. Fukui T, Tabata M, Matsuyama S, Takanashi S. Graft selection in elderly patients undergoing coronary artery bypass grafting. Gen Thorac Cardiovasc Surg. 2011;59:786–92. Lytle BW, Blackstone EH, Sabik JF, Houqhtalinq P, Loop FD, Cosgrove DM. The effect of bilateral internal thoracic artery grafting on survival during 20 postoperative years. Ann Thorac Surg. 2004;78:2005–14. Serruys BW, Morice MC, Kappetein AP, et al. Percutaneous coronary intervention versus coronary-artery bypass for severe coronary artery disease. N Engl J Med. 2009;360(10):964–72. Biancari F, Mosorin M, Rasinaho E, et al. Postoperative stroke after off-pump versus on-pump coronary artery bypass surgery. J Thorac Cardiovasc Surg. 2007;133:169–73. Li Z, Denton T, Yeo KK, et al. Off-pump bypass surgery and postoperative stroke: California coronary bypass outcomes reporting program. Ann Thorac Surg. 2010;90:753–9. Banerjee A, Fowkes FG, Rothwell PM. Associations between peripheral artery disease and ischemic stroke: implications for primary and secondary prevention. Stroke. 2010;41:2102–7.
123