Off-pump versus on-pump coronary artery bypass surgery in patients with actively treated diabetes and multivessel coronary disease

Accepted Manuscript Off-pump versus on-pump coronary artery bypass surgery in patients with actively treated diabetes and multivessel coronary disease Umberto Benedetto, PhD, Massimo Caputo, MD, Hunaid Vohra, PhD, Alan Davies, James Hillier, MD, Alan Bryan, MD, Gianni D. Angelini, MD. PII:

S0022-5223(16)30668-7

DOI:

10.1016/j.jtcvs.2016.06.038

Reference:

YMTC 10694

To appear in:

The Journal of Thoracic and Cardiovascular Surgery

Received Date: 19 February 2016 Revised Date:

23 May 2016

Accepted Date: 17 June 2016

Please cite this article as: Benedetto U, Caputo M, Vohra H, Davies A, Hillier J, Bryan A, Angelini GD, Off-pump versus on-pump coronary artery bypass surgery in patients with actively treated diabetes and multivessel coronary disease, The Journal of Thoracic and Cardiovascular Surgery (2016), doi: 10.1016/ j.jtcvs.2016.06.038. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT 1

Off-pump versus on-pump coronary artery bypass surgery in patients with

2

actively treated diabetes and multivessel coronary disease.

3 4 Umberto Benedetto PhD, Massimo Caputo MD, Hunaid Vohra PhD, Alan Davies,

6

James Hillier MD, Alan Bryan MD, Gianni D Angelini MD.

7

RI PT

5

Bristol Heart Institute, University of Bristol, School of Clinical Sciences, Bristol, UK

9

No potential conflicts exist for all authors

SC

8

This study was supported by the British Heart Foundation and the NIHR Bristol

11

Cardiovascular Biomedical Research Unit

12

Word count: 2908

13

M AN U

10

Corresponding Author

15

Umberto Benedetto MD PhD

16

Bristol Heart Institute, University of Bristol

17

Upper Maudlin St, Bristol BS2 8HW

18

Tel: +44 (0) 117 3422856

19

Email: [email protected]

AC C

EP

TE D

14

Abbreviations

21

AF: atrial fibrillation

22

BITA: bilateral internal thoracic artery

23

BMI: body mass index

24

CCS: Canadian Cardiovascular Society

25

CVA: cerebrovascular event

26

COPD: chronic obstructive pulmonary disease;

27

DM: diabetes

28

IR: incomplete revascularization

29

IABP intra-aortic balloon pump

30

LMD: left main disease

31

LVEF: left ventricular ejection fraction

32

MI: myocardial infarction

33

OPCAB: off-pump coronary artery bypass

34

ONCAB: on-pump coronary artery bypass

35

PCI: percutaneous coronary intervention

36

PVD: peripheral vascular disease

37

RRT: renal replacement therapy

38

sCr: serum creatinine

39

TVD: three vessel disease

40

PSM: propensity score matching

41

SDM: standardized mean difference

AC C

EP

TE D

M AN U

SC

20

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT Central message

43

Off-pump bypass surgery is a safe and feasible option for diabetic patients with

44

multivessel disease, it reduces the incidence of early complications and provides

45

long term survival benefit similar to on-pump surgery

AC C

EP

TE D

M AN U

SC

RI PT

42

ACCEPTED MANUSCRIPT Perspective statement

47

Off-pump bypass surgery remains valid option for diabetic patients with multivessel

48

disease as it reduces the incidence of early complications and provides long term

49

survival benefit similar to on-pump surgery. Its use should not be discouraged in

50

such a high risk subgroup.

AC C

EP

TE D

M AN U

SC

RI PT

46

ACCEPTED MANUSCRIPT Abstract

52

Objective(s): We conducted a single centre analysis on short-term outcomes and

53

long term survival in actively treated diabetic patients undergoing off pump coronary

54

artery bypass (OPCAB) versus on pump coronary artery bypass (OPCAB) surgery.

55

Methods: The final population consisted of 2450 patients with actively treated

56

diabetes (mean age 66±9 years; F/M 545/1905, 22%). Of those, 1493 subjects were

57

orally treated and 1011 were on insulin. OPCAB and ONCAB were performed in

58

1253 and 1197 patients respectively. Propensity score matching was used to

59

compare the two matched groups.

60

Results: When compared to ONCAB, OPCAB was associated with a significant risk

61

reduction for postoperative cerebrovascular accident (OR 0.49; 95%CI 0.25-0.99;

62

P=0.04), need for postoperative IABP (OR 0.48; 95%CI 0.30-0.77; P=0.002) and re-

63

exploration for bleeding (OR 0.55; 95%CI 0.33-0.94; P=0.02). OPACB did not

64

significantly affect early (HR 1.32; 95%CI 0.73-2.40; P=0.36) and late mortality (HR

65

1.08;95%CI 0.92-1.28; P=0.32). However, OPCAB with incomplete revascularization

66

was associated with a reduced survival rate when compared to OPCAB with

67

complete revascularization (HR 1.82; 95%CI 1.34-2.46; P=0.0002) and ONCAB with

68

complete revascularization (HR 1.83; 95%CI1.36-2.47; P<0.0001).

69

Conclusions: OPCAB is a safe and feasible option for diabetic patients with

70

multivessel disease, it reduces the incidence of early complications including

71

postoperative cerebrovascular events and provides excellent long term survival

72

similar to ONCAB surgery in case of complete revascularization.

73

Word count: 219

AC C

EP

TE D

M AN U

SC

RI PT

51

ACCEPTED MANUSCRIPT Diabetes mellitus (DM) is a major public health and economic problem with a

75

dramatic increase in prevalence and incidence [1]. Coronary heart disease is highly

76

prevalent and is the major cause of morbidity and mortality among diabetic patients

77

[2]. Patients with DM account for approximately one quarter of all patients who

78

undergo coronary revascularization procedures each year [3]. Coronary artery

79

bypass graft (CABG) surgery when compared with percutaneous coronary

80

intervention (PCI) has been shown to be associated with better outcomes in patients

81

with diabetes [4], and current clinical guidelines recommend CABG as the preferred

82

revascularization strategy in diabetic patients with complex coronary artery disease

83

[5]. However, DM has a significant negative impact on the clinical outcome of

84

coronary surgery in the long term as well as in the short term [6]. The debate on the

85

effectiveness or otherwise of off pump coronary artery bypass (OPCAB) versus on

86

pump coronary artery bypass (ONCAB) surgery continues [7-9]. In particular the role

87

of OPCAB in the management of diabetic patients remains controversial as

88

comparative data are scarce and inconclusive [10-13]. A recent sub-analysis of the

89

Bypass Angioplasty Revascularization Investigation 2 Diabetes trial [10] suggested

90

that patients with diabetes had greater risk of major cardiovascular long-term events

91

after OPCAB than after ONCAB. However, this study presented several limitations

92

including very small sample size and definitive conclusions could not be drawn [14].

93

In this study, we used a large consecutive cohort with DM operated over a 20 years

94

period at a single institution, to perform a propensity score matching (PSM) analysis

95

of ONCAB versus OPCAB on short-term outcomes and long term survival.

96

Methods

97

The study was conducted in accordance with the principles of the Declaration of

98

Helsinki. The local audit committee approved the study, and the requirement for

AC C

EP

TE D

M AN U

SC

RI PT

74

ACCEPTED MANUSCRIPT individual patient consent was waived. We retrospectively analysed prospectively

100

collected data from The National Institute for Cardiovascular Outcomes Research

101

(NICOR) NACSA registry for all isolated first time CABG procedures performed at

102

the Bristol Heart Institute, Bristol United Kingdom from April 1996 to April

103

2015. Reproducible cleaning algorithms were applied to the database, which are

104

regularly updated as required. Briefly, duplicate records and non-adult cardiac

105

surgery entries were removed; transcriptional discrepancies harmonized; and clinical

106

conflicts and extreme values corrected or removed. The data are returned regularly

107

to the local units for validation.

108

Further details and definition of variables are available at

109

http://www.ucl.ac.uk/nicor/audits/adultcardiac/datasets. From 15119 isolated first

110

time CABG cases performed during the study period, we selected subjects with

111

multivessel coronary artery disease and actively treated DM at the time of surgery.

112

Patients with no diabetes (n=12439), single vessel disease (n=121) and no data

113

regarding surgical strategy adopted (OPCAB versus ONCAB, n=109) were excluded.

114

The final population consisted of 2450 patients (22%, mean age 66±9 years; F/M

115

545/1905) with actively treated diabetes. Of those, 1493 patients were orally treated

116

and 1011 were on insulin. OPCAB (Video 1) and ONCAB were performed in 1253

117

and 1197 patients respectively (Figure 1) according to surgeon preference and not

118

based on specific clinical indications. Normothermic blood cardioplegia was the

119

standard strategy for ONCAB cases during the study period. Blood glucose level was

120

strictly maintained <11.1 mmol/L perioperatively according to unit protocol, using

121

insulin infusion.

AC C

EP

TE D

M AN U

SC

RI PT

99

122 123

Pre-treatment variables and study end-points

ACCEPTED MANUSCRIPT The effect of OPCAB over ONCAB was adjusted for the following variables including:

125

age, gender, body mass index (BMI); previous myocardial infarction (MI) within 30

126

days, previous percutaneous coronary intervention (PCI); diabetes mellitus (DM) on

127

insulin; chronic obstructive pulmonary disease (COPD); Canadian Cardiovascular

128

Society (CCS) class III or IV; New York Heart Association (NYHA) class III or IV;

129

current smoking; serum creatinine ≥200 mmol/l, previous cerebrovascular accident

130

(CVA); peripheral vascular disease (PVD); preoperative atrial fibrillation (AF); left

131

main disease (LMD); three vessel disease (TVD); left ventricle ejection fraction

132

(LVEF) ≤49% and ≤30%, pre-operative use of intra-aortic balloon pump (IABP), non-

133

elective admission, emergent/salvage operation, cardiogenic shock and year of

134

operation. Overall risk profile was evaluated by using additive and logistic Euroscore.

135

Incomplete revascularization (IR) was defined as at least one diseased primary

136

arterial territory not grafted.

137

Short term outcomes investigated were postoperative complications including CVA,

138

need for IABP defined as unplanned insertion of IABP intraoperatively or

139

postoperatively due to hemodynamic instability, re-exploration for bleeding, renal

140

replacement therapy (RRT), sternal wound reconstruction as single and combined

141

end-points and mortality within 30 days. Long term outcome was all-cause mortality.

142

All-cause mortality is the most robust and unbiased index because no adjudication is

143

required; thus, inaccurate or biased documentation or clinical assessments are

144

avoided [15]. Information about death was obtained from the institutional database

145

and the National General Register Office for all patients. Data regarding

146

postoperative complications and survival were available for all patients in the study.

147

Statistical analysis

AC C

EP

TE D

M AN U

SC

RI PT

124

ACCEPTED MANUSCRIPT For baseline characteristics, variables are summarized as mean for continuous

149

variables and proportion for categorical variables. Multiple imputation using

150

bootstrapping-based expectation-maximization algorithm and including all pre-

151

treatment variables (Amelia R package, http://www.jstatsoft.org/v45/i07/) was used

152

to address missing data (Supplementary Table 1). To control for measured potential

153

confounders in the data set, a propensity score (PS) was generated for each patient

154

from a multivariable logistic regression model based on pre-treatment covariates as

155

independent variables with treatment type (OPCAB vs ONCAB) as a binary

156

dependent variable [16]. The resulting propensity score represented the probability

157

of a patient to undergo OPCAB (Area under the curve 0.63, Supplementary Figure

158

2). Pairs of patients receiving OPCAB and ONCAB were derived using greedy 1:1

159

matching with a calliper of width of 0.2 standard deviation of the logit of the PS (non-

160

random R package, http://CRAN.Rproject.org/package=nonrandom). The quality of

161

the match was assessed by comparing selected pre-treatment variables in

162

propensity score– matched patient using the standardized mean difference (SMD),

163

by which an absolute standardized difference of greater than 10% is suggested to

164

represent meaningful covariate imbalance [16]. Analytic methods for the estimation

165

of the treatment effect in the matched sample included McNemar’s to compare

166

proportions [16]. Time-segmented Cox regression models (within 30 days and

167

beyond 30 days) [17] that stratified on the matched pairs [18] were used to

168

investigate the effect of treatment (ONCAB vs ONCAB) on early and late mortality.

169

This approach accounts for the within-pair homogeneity by allowing the baseline

170

hazard function to vary across matched sets (survival R package, http://CRAN.R-

171

project.org/package=survival). Lastly a multivariate adjustment for all baseline

172

characteristics and intraoperative variables such as the use of bilateral internal

AC C

EP

TE D

M AN U

SC

RI PT

148

ACCEPTED MANUSCRIPT thoracic arteries (BITAs), Radial Artery (RA) and the incidence of incomplete

174

revascularization was performed in the matched sample to correct the effect of

175

OPCAB for residual imbalance (double robust) and to estimate the effect size of

176

other covariates on all outcomes investigated. Person correlation coefficient and

177

variance inflation factors (VIF) from the covariance matrix of parameter estimates

178

was used to assess collinearity in multivariate models (usdm R package,

179

http://CRAN.R-project.org/package=usdm). Variables with VIF higher than 4 were

180

excluded from multivariate analyses. All p-values <0.05 were considered to indicate

181

statistical significance. All statistical analysis was performed using R Statistical

182

Software (version 3.2.3; R Foundation for Statistical Computing, Vienna, Austria).

183

Results

184

Pre-treatment variables distribution in OPCAB and ONCAB groups are summarized

185

in Table 1. Patients undergoing ONCAB were likely to have CCS III/IV, three vessel

186

disease, LVEF≤30% and cardiogenic shock. Era of surgery was also different

187

between the two groups. Additive Euroscore (P=0.02) but not logistic Euroscore

188

(P=0.15) was higher in the ONCAB group. PSM created a total of 995 pairs perfectly

189

matched for all pre-treatment variables (SMD<0.10) including era of surgery. In the

190

matched population additive Euroscore (P=0.08) and logistic Euroscore (P=0.12)

191

were comparable between the two groups.

192

Intraoperative data

193

Table 2 summarizes intraoperative data. When compared to ONCAB, OPCAB was

194

associated with a lower number of grafts per patient (2.7±0.7 versus 3.0±0.7

195

P<0.001; Figure 2) in the right and circumflex but not the left anterior descending

196

artery territory. The rate of BITA was particular low in both groups (P=0.1). The use

197

of the RA was higher in the OPCAB group while saphenous vein graft more

AC C

EP

TE D

M AN U

SC

RI PT

173

ACCEPTED MANUSCRIPT frequently used in the ONCAB group. OPCAB conversion rate to ONCAB

was

199

7/1253 (0.6%).

200

Effect of OPCAB on short term outcomes

201

Short term outcomes are summarized in Table 3. OPCAB was associated with a

202

higher rate of IR. When compared to ONCAB, OPCAB was significantly associated

203

with 50% relative risk reduction for postoperative CVA (P=0.04), need for

204

postoperative IABP (P=0.002) and re-exploration for bleeding (P=0.02). Overall 30

205

days mortality was 2.4% (58 pts) with no significant difference between the two

206

groups. This trend toward a benefit from OPCAB on short term outcomes was

207

confirmed when results were analysed separately for patients on oral or insulin

208

treatment (Supplementary Table 2 and 3). There were no death in the seven OPCAB

209

patients who required conversion to ONCAB.

210

Effect of OPCAB on long term mortality

211

At a mean follow-up of 6.5±4.5 years, there were 357 and 408 deaths in the

212

unmatched population and 284 and 299 deaths in the matched sample in the

213

OPCAB and ONCAB groups respectively (Figure 2). In the matched sample survival

214

probability at 5 and 10 was 82.6±1.2% and 62.6±2.0% versus 84.3±1.3% and

215

64.0±1.9% in the OPCAB and ONCAB group respectively. The two strategies were

216

perfectly comparable in terms of late mortality in the overall matched sample (HR

217

1.08;95%CI

218

1.06;95%CI0.85-1.33; P=0.6) and on insulin (HR 1.12; 95%CI 0.88-1.43; P=0.35,

219

Figure 3).

220

OPCAB with complete revascularization showed comparable survival when

221

compared to ONCAB with complete revascularization (n=921) (HR 1.00; 95%CI

222

0.84-1.20; P=0.96).

AC C

EP

TE D

M AN U

SC

RI PT

198

0.92-1.28;

P=0.32)

and

among

patients

orally

treated

(HR

However, OPCAB patients with IR (n=139) showed reduced

ACCEPTED MANUSCRIPT survival rate when compared to OPCAB with complete revascularization (n=856)

224

(HR 1.82; 95%CI 1.34-2.46; P=0.0002) and ONCAB with complete revascularization

225

(HR 1.83; 95%CI1.36-2.47; P<0.0001; Figure 4). We could not draw conclusion on

226

the impact on IR among ONCAB subjects due to the small sample size (n=74).

227

Effect of OPCAB over ONCAB after full adjustment (double robust)

228

After double robust adjustment, OPCAB was confirmed to be significantly associated

229

with a reduced incidence of postoperative stroke, need for postoperative IABP,

230

sternal wound reconstruction, re-exploration for bleeding but not dialysis. OPACB

231

was

232

revascularization. OPACB was not associated with increased early (within 30 days)

233

and late mortality (Supplementary Table 4, central picture).

234

Impact of other predictors of outcomes

235

Impact of other predictors on outcomes of interest is reported in Supplementary

236

Tables 5 to 13. In particular, diabetes on insulin was significantly associated with

237

increased risk of postoperative dialysis and incidence of any complication. Diabetes

238

on insulin did not significantly affect early mortality but was associated with 18%

239

relative risk increase in late mortality. IR did not significantly influence operative

240

outcomes and early mortality, but it was associated with nearly 50% relative risk

241

increase in late mortality. All variables investigated showed VIF < 4 (Supplementary

242

Table 14).

243

Discussion

244

To the best of our knowledge this is the largest series with the longest follow-up

245

reported on the impact of OPCAB in diabetic patients with multivessel coronary

246

disease. OPCAB was associated with a trend towards a lower incidence of

247

postoperative complications including cardiovascular accidents, and comparable

associated

with

a

2-fold

increased

risk

of

incomplete

AC C

EP

TE D

M AN U

significantly

SC

RI PT

223

ACCEPTED MANUSCRIPT early and late survival when compared to ONCAB. The present analysis confirmed

249

that OPCAB was associated with a ~7% absolute risk increase in incomplete

250

revascularization. Furthermore, OPCAB with IR showed a lower survival rate when

251

compared to OPCAB and ONCAB with complete revascularization. Insulin treatment

252

was confirmed to be an independent risk factor for postoperative complications and

253

poorer long term survival.

254

Despite several randomized controlled trials there is still controversy with respect to

255

the advantages and disadvantages of OPCAB vs OPCAB surgery [7-9,19,20],

256

However, some evidence has suggested a potential off-pump benefit for high

257

operative risk patient subgroups [21-23]. In the case of diabetic patients with

258

multivessel disease only limited data are available with conflicting reports [10-13]. A

259

recent observational sub-analysis of the BARI 2D trial [10] concluded that patients

260

with diabetes had greater risk of major long-term cardiovascular events after OPCAB

261

than after ONCAB. However, this study has been criticized [14] due to the small

262

number of patients compared (153 propensity matched pairs) which is inadequate to

263

draw final conclusions. Moreover there was missing information regarding the

264

completeness of revascularization. A sub-analysis of the ROOBY trial [11] on

265

diabetic patients randomized to OPCAB (n=402) or ONCAB (n=433) concluded that

266

1-year graft patency rate was lower after OPCAB with no difference in the 1-year

267

primary composite outcome. However, it is largely recognized that in the ROOBY

268

Trial, surgeons were inexperienced (on average performing eight OPCAB operations

269

per year and with a high conversion rate of around 12% [14]. The importance of

270

surgeon experience in OPCAB is clearly demonstrated in trials that showed no

271

difference in hard clinical end-points at 1 year or at longer follow up [7,8,24].

272

Moreover, the 1% lower graft patency rate reported in the OPCAB group is clearly

AC C

EP

TE D

M AN U

SC

RI PT

248

ACCEPTED MANUSCRIPT marginal (83.1% vs 88.4%) and it is likely not to have any clinical impact. On the

274

other hand, Emmert et al. [12] compared short term outcomes in 540 OPCAB and

275

475 ONCAB diabetic patients, and showed a clear trend to reduced major

276

complications such as stroke, re-thoracotomy for bleeding, and postoperative IABP

277

requirement, confirming the beneficial effect of OPCAB on short term outcomes in

278

this subset of patients. Renner et al. [13] compared 355 OPCAB and 502 ONCAB

279

procedure and they concluded that OPCAB was associated with a significant lower

280

30-day mortality risk and also with a lower 1 year mortality risk compared with

281

ONCAB.

282

In the present analysis, IR occurred in 15.2% and 7.9% of unmatched OPCAB and

283

ONCAB groups respectively. Although patients undergoing OPCAB with complete

284

revascularization showed survival rates similar to those receiving ONCAB, OPCAB

285

with IR was found to be significantly associated with a lower survival rate when

286

compared to OPCAB and ONCAB with complete revascularization. It has been

287

suggested that patients who undergo IR have multiple comorbidities and unfavorable

288

anatomy that could bias the data in favour of complete revascularization [25].

289

However compelling evidence supports the hypothesis that diabetics are more likely

290

to benefit from a more complete revascularization [26]. Raza et al. recently reported

291

a long term survival analysis on 11922 diabetic patients [27] and demonstrated that

292

complete revascularization was associated with 10% lower late mortality. Moreover,

293

they found that OPCAB was associated with 10% lower late mortality than ONCAB,

294

but the difference was not statistically significant (P = 0.2). Nakano et al. [28]

295

reported on 604 consecutive patients underwent OPCAB during a 6-year period with

296

216 patients receiving IR (13%), They found that all the event-free survival rates for

297

all-cause mortality (P < .001), cardiac death (P = .020), and major adverse cardiac

AC C

EP

TE D

M AN U

SC

RI PT

273

ACCEPTED MANUSCRIPT and cerebrovascular events (P < .001) were lower in the IR group. Of note, IR rate

299

observed in the present OPCAB series was significantly lower than that reported by

300

others. Omer at al. [29] recently reported a 29% rate of IR in 6367 OPCAB cases

301

compared to 11.0% in 34,772 ONCAB cases. Kleisli et al. [30] reported a 22% rate of

302

IR in 207 OPCAB and 6.3% in 827 ONCAB patients. Some have suggested that

303

surgeon experience plays a role and emphasized the importance of surgeon volume

304

in establishing favorable outcomes for OPCAB [24]. The relatively low rate of IR in

305

the present OPCAB series is likely to be related to the high OPCAB volume in our

306

centre and support the hypothesis that when OPCAB is routinely performed,

307

complete revascularization can be achieved in the majority of cases. The trend

308

toward shorter survival with OPCAB observed in some studies may be related to

309

incomplete revascularization [27,28], more likely to happened with surgeons less

310

experienced with this technique. Finally, although the use of additional arterial grafts

311

has been shown to minimize the risk related to incomplete revascularization [31], in

312

the present series the rate of bilateral internal thoracic arteries usage was low (~2%).

313

This may represent the concern of increased sternal wound complication in this high

314

risk group of patients [27].

315

Limitations

316

The main limitation of our study is that no follow-up data were available to compare

317

the groups with respect to the cause of death (cardiac vs non-cardiac), recurrence of

318

angina, need for repeated revascularization, and graft patency. Therefore, we can

319

only speculate about the mechanism beyond the equipoise between OPCAB and

320

ONCAB on long-term survival. Finally, although the data were collected

321

prospectively, the nonspecific design for the diabetic population limits the present

322

analysis. Propensity technique can adjust only for measurable and included

AC C

EP

TE D

M AN U

SC

RI PT

298

ACCEPTED MANUSCRIPT variables but cannot exclude a selection bias based on non-measurable “eye-ball”. In

324

conclusion, OPCAB is a safe and feasible option for diabetic patients with

325

multivessel coronary disease. OPCAB reduces the incidence of postoperative

326

complications and provides long term survival comparable to ONCAB.

327

Complete revascularization should still be the main goal while performing OPCAB

328

surgery in diabetic patients.

RI PT

323

AC C

EP

TE D

M AN U

SC

329

ACCEPTED MANUSCRIPT 330

References

331

1) Narayan KMV, Gregg EW, Fagot-Campagna A, Engelgau MM, Vinicor F.

332

Diabetes: a common, growing, serious, costly, and potentially preventable public

333

health problem. Diabetes Res Clin Pract 2000;50(Suppl):77– 84. 2) Berry C, Tardif JC, Bourassa MG. Coronary heart disease in patients with

335

diabetes. Part I: recent advances in prevention and noninvasive management. J

336

Am Coll Cardiol 2007;49:631–42.

RI PT

334

3) Berry C, Tardif JC, Bourassa MG. Coronary heart disease in patients with

338

diabetes. Part II: recent advances in coronary revascularization. J Am Coll

339

Cardiol 2007;49:643–56.

M AN U

SC

337

340

4) Farkouh ME, Domanski M, Sleeper LA, Siami FS, Dangas G, Mack M et al.

341

Strategies for multivessel revascularization in patients with diabetes. N Engl J

342

Med 2012;367:2375–84.

5) Kolh P, Windecker S, Alfonso F, Collet J-P, Cremer J, Falk V et al. 2014 ESC/

344

EACTS Guidelines on myocardial revascularization: The Task Force on

345

Myocardial Revascularization of the European Society of Cardiology (ESC) and

346

the European Association for Cardio-Thoracic Surgery (EACTS). Eur J

347

Cardiothorac Surg 2014;46:517–92

AC C

EP

TE D

343

348

6) Kappetein AP, Head SJ,Morice MC, Banning AP, Serruys PW, Mohr FW et al.

349

Treatment of complex coronary artery disease in patients with diabetes: 5-year

350

results comparing outcomes of bypass surgery and percutaneous coronary

351

intervention in the SYNTAX trial. Eur J Cardiothorac Surg 2013;43:1006–13.

352

7) Lamy A, Devereaux PJ, Prabhakaran D, Taggart DP, Hu S, Paolasso E et al.

353

Effects of off-pump and on-pump coronary-artery bypass grafting at 1 year. N

354

Engl J Med 2013;368:1179–88.

ACCEPTED MANUSCRIPT 355

8) Diegeler A, Börgermann J, Kappert U, Breuer M, Böning A, Ursulescu A et al.

356

Off-pump versus on-pump coronary-artery bypass grafting in elderly patients. N

357

Engl J Med 2013;368:1189–98. 9) Hattler B, Messenger JC, Shroyer AL, Collins JF, Haugen SJ, Garcia JA et al.

359

Off-Pump coronary artery bypass surgery is associated with worse arterial and

360

saphenous vein graft patency and less effective revascularization: results from

361

the Veterans Affairs Randomized On/Off Bypass (ROOBY) trial. Circulation

362

2012;125:2827–35.

SC

RI PT

358

10) Singh A, Schaff HV, Mori Brooks M, Hlatky MA, Wisniewski SR, et al; BARI 2D

364

Study Group On-pump versus off-pump coronary artery bypass graft surgery

365

among patients with type 2diabetes in the Bypass Angioplasty Revascularization

366

Investigation 2 Diabetes trial. Eur J Cardiothorac Surg. 2016;49:406-16.

367

M AN U

363

11) Shroyer AL, Hattler B, Wagner TH, Baltz JH, Collins JF, Carr BM et al. VA #517 Randomized

On/Off Bypass (ROOBY)

369

pump and on-pump clinical

370

surgery patients. Ann Thorac Surg. 2014;98:38-44

TE D

368

outcomes

Study

and

Group.

costs

for

Comparing offdiabetic

cardiac

12) Emmert MY, Salzberg SP, Seifert B, Rodriguez H, Plass A, Hoerstrup SP et al. Is

372

off-pump superior to conventional coronary artery bypass grafting in diabetic

373

patients with multivessel disease? Eur J Cardiothorac Surg. 2011;40:233-9.

AC C

374

EP

371

13) Renner A, Zittermann A, Aboud A, Pühler T, Hakim-Meibodi K, Quester W et al.

375

Coronary revascularization

in

diabetic

376

pump surgery. Ann Thorac Surg. 2013;96:528-34

patients: off-pump versus on-

377

14) Taggart DP. Off-pump coronary artery bypass graft in patients with type 2

378

diabetes: pushing the Bypass Angioplasty Revascularization Investigation Type 2

379

Diabetes trial too far. Eur J Cardiothorac Surg. 2016;49:416-8

ACCEPTED MANUSCRIPT 380 381

15) Lauer MS, Blackstone EH, Young JB, Topol EJ. Cause of death in clinical research: time for a reassessment? J Am Coll Cardiol. 1999;34:618–20 16) Austin PC. A tutorial and case study in propensity score analysis: an application

383

to estimate the effect of in-hospital smoking cessation counseling on mortality.

384

Multivariate Behav Res, 2011;46:119–151

385

17) Myers WO, Blackstone EH, Davis K, Foster ED, Kaiser GC. CASS Registry long term

387

Cardiol. 1999;33:488-98.

survival.

Coronary

Artery

Surgery Study.

J

Am

Coll

18) Cummings P. McKnight B. Greenland S. Matched cohort methods for injury research. Epidemiologic Reviews. 2003;25:43–50

M AN U

389

surgical

SC

386

388

RI PT

382

390

19) Angelini GD, Taylor FC, Reeves BC, Ascione R. Early and midterm outcome after off-pump

391

and on-pump surgery in Beating Heart Against Cardioplegic Arrest Studies (BHACAS 1 and

392

2): a pooled analysis of two randomised controlled trials. Lancet 2002;359:1194-1199. 20) Ascione R, Reeves BC, Taylor FC, Seehra HK, Angelini GD. Beating heart against

TE D

393 394

cardioplegic arrest studies (BHACAS 1 and 2): quality of life at mid-term follow-up in two

395

randomised controlled trials. Eur Heart J 2004;25:765-770 21) Cavallaro P, Itagaki S, Seigerman M, Chikwe J. Operative mortality and stroke

397

after on-pump vs off-pump surgery in high-risk patients: an analysis of 83,914

398

coronary bypass operations. Eur J Cardiothorac Surg 2014;45:159-64

AC C

EP

396

399

22) Marui A, Okabayashi H, Komiya T, Tanaka S, Furukawa Y, Kita T et al.; CREDO-

400

Kyoto Investigators. Benefits of off-pump coronary artery bypass grafting in high-

401

risk patients. Circulation. 2012 Sep 11;126(11 Suppl 1):S151-7

402

23) Ascione R, Reeves BC, Rees K, Angelini GD. Effectiveness of coronary artery

403

bypass grafting with or without cardiopulmonary bypass in overweight patients.

404

Circulation. 2002;106:1764-70.

ACCEPTED MANUSCRIPT 405

24) Lapar DJ, Mery CM, Kozower BD, Kern JA, Kron IL, Stukenborg GJ et al. The

406

effect

of

surgeon

volume

on mortality for off-pump coronary

407

bypass grafting. J Thorac Cardiovasc Surg. 2012;143:854-63.

artery

25) Gossl M, Faxon DP, Bell MR, Holmes DR, Gersh BJ. Complete versus

409

incomplete revascularization with coronary artery bypass graft or percutaneous

410

intervention in stable coronary artery disease. Circ Cardiovasc Interv. 2012;5:

411

597-604.

26) Takagi H, Watanabe T, Mizuno Y, Kawai N, Umemoto T; ALICE (All-Literature

413

Investigation

Cardiovascular

Evidence)

Group.

A

meta-analysis

414

of adjusted risk estimates for survival from observational studies of complete

415

versus incomplete revascularization in patients with multivessel disease

416

undergoing coronary artery bypass grafting. Interact Cardiovasc Thorac Surg.

417

2014;18:679-82.

M AN U

of

SC

412

RI PT

408

27) Raza S, Sabik JF 3rd, Masabni K, Ainkaran P, Lytle BW, Blackstone EH. Surgical

419

revascularization techniques that minimize surgical risk and maximize late

420

survival after coronary artery bypass grafting in patients with diabetes mellitus. J

421

Thorac Cardiovasc Surg. 2014;148:1257-1264

EP

TE D

418

28) Nakano J, Okabayashi H, Noma H, Sato T, Sakata R. The impact of incomplete

423

revascularization and angiographic patency on midterm results after off-pump

424

coronary artery bypass grafting. J Thorac Cardiovasc Surg. 2014;147:1225-32.

425

29) Omer S, Cornwell LD, Rosengart TK, Kelly RF, Ward HB, Holman WL et al.

426

Completeness of coronary revascularization and survival: Impact of age and off-

427

pump surgery. J Thorac Cardiovasc Surg. 2014;148:1307-15

AC C

422

ACCEPTED MANUSCRIPT 428

30) Kleisli T, Cheng W, Jacobs MJ, Mirocha J, Derobertis MA, Kass RM et al.In

429

the current era, complete revascularization improves survival after coronary arter

430

y bypass surgery. J Thorac Cardiovasc Surg. 2005;129:1283-9 31) Kieser TM, Curran HJ, Rose MS, Norris CM, Graham MM. Arterial grafts balance

432

survival between incomplete and complete revascularization: a series of 1000

433

consecutive coronary artery bypass graft patients with 98% arterial grafts. J

434

Thorac Cardiovasc Surg. 2014;147:75-83.

AC C

EP

TE D

M AN U

SC

RI PT

431

ACCEPTED MANUSCRIPT

Table 1. Pre-treatment variables distribution in the unmatched and matched OPCAB and ONCAB groups

No Yes

976 277

77.9 22.1

929 268

BMI

<30 ≥30

723 530

57.7 42.3

726 471

CCS

No Yes

654 599

52.2 47.8

NYHA

No Yes

791 462

63.1 36.9

MI within 30 days

No Yes

985 268

PCI

No Yes

DM on insulin

No Yes

PS-matched OPCAB n % 995 100.0 242 24.3 360 36.2 346 34.8 47 4.7

PS-matched OPCAB n % 995 100.0 233 23.4 393 39.5 345 34.7 24 2.4

RI PT

Female

4.8

SMD

2.4

0.7

765 230

76.9 23.1

762 233

76.6 23.4

0.7

60.7 39.3

7.5

575 420

57.8 42.2

575 420

57.8 42.2

1.1

43.9 56.1

16.5

514 481

51.7 48.3

491 504

49.3 50.7

4.6

704 493

58.8 41.2

8.8

624 371

62.7 37.3

611 384

61.4 38.6

2.7

78.6 21.4

932 265

77.9 22.1

1.8

796 199

80.0 20.0

768 227

77.2 22.8

6.8

1157 96

92.3 7.7

1125 72

94.0 6.0

6.5

917 78

92.2 7.8

925 70

93.0 7.0

3.0

733 520

58.5 41.5

706 491

59.0 41.0

0.9

583 412

58.6 41.4

584 411

58.7 41.3

0.2

TE D

77.6 22.4

526 671

EP

Total sample Age

SMD

SC

Unmatched OPCAB n % 1197 100.0 282 23.6 490 40.9 401 33.5 24 2.0

M AN U

<60.0 60-69 70-79 ≥8

Unmatched OPCAB n % 1253 100.0 307 24.5 449 35.8 440 35.1 57 4.5

AC C

435

ACCEPTED MANUSCRIPT

No Yes

1117 136

89.1 10.9

1060 137

88.6 11.4

1.8

889 106

89.3 10.7

888 107

89.2 10.8

0.3

sCr>200mmol/l

No Yes

1204 49

96.1 3.9

1147 50

95.8 4.2

1.3

956 39

96.1 3.9

958 37

96.3 3.7

1.0

COPD

No Yes

1140 113

91.0 9.0

1086 111

90.7 9.3

0.8

903 92

90.8 9.2

899 96

90.4 9.6

1.3

CVA

No Yes

1179 74

94.1 5.9

1131 66

94.5 5.5

931 64

93.6 6.4

940 55

94.5 5.5

3.8

PVD

No Yes

1075 178

85.8 14.2

991 206

AF

No Yes

1198 55

95.6 4.4

1142 55

TVD

No Yes

346 907

27.6 72.4

LMD

No Yes

964 289

76.9 23.1

LVEF<50%

No Yes

839 414

LVEF 30-49%

No Yes

LVEF<30%

No Yes

SC

RI PT

Current smoking

M AN U

1.6

8.2

853 142

85.7 14.3

848 147

85.2 14.8

1.4

95.4 4.6

0.9

949 46

95.4 4.6

951 44

95.6 4.4

0.9

19.2 80.8

19.9

298 697

29.9 70.1

224 771

22.5 77.5

9.2

913 284

76.3 23.7

1.5

770 225

77.4 22.6

753 242

75.7 24.3

4.0

67.0 33.0

786 411

65.7 34.3

2.7

660 335

66.3 33.7

663 332

66.6 33.4

0.6

923 330

73.7 26.3

902 295

75.4 24.6

2.7

729 266

73.3 26.7

737 258

74.1 25.9

1169 84

93.3 6.7

1081 116

90.3 9.7

10.9

926 69

93.1 6.9

921 74

92.6 7.4

TE D

82.8 17.2

AC C

EP

230 967

1.9

ACCEPTED MANUSCRIPT

No Yes

1247 6

99.5 0.5

1178 19

98.4 1.6

10.9

990 5

99.5 0.5

990 5

99.5 0.5

0.0

preop IABP

No Yes

1237 16

98.7 1.3

1167 30

97.5 2.5

9.0

981 14

98.6 1.4

981 14

98.6 1.4

0.3

non elective

No Yes

644 609

51.4 48.6

560 637

46.8 53.2

9.2

520 475

52.3 47.7

476 519

47.8 52.2

8.8

Emergent/salvage

No Yes

1239 14

98.9 1.1

1171 26

97.8 2.2

982 13

98.7 1.3

980 15

98.5 1.5

1.7

Era of surgery

19962004 20052007 20082010 20112015

398

31.8

306

254

25.5

256

25.7

9.8

267

21.3

349

29.2

273

27.4

276

27.7

324

25.9

283

23.6

240

24.1

269

27.0

264

21.1

259

21.6

228

22.9

194

19.5

Logistic Euroscore

5±5%

SC 8.3

M AN U 25.6

TE D

EP

4.2±2.6

AC C

AdditiveEuroscore

RI PT

Cardiogenic shock

34.6

4.4±2.8

4.2±2.7

4.2±2.5

5±7%

5±5%

4±5%

436

OPCAB: off-pump coronary artery bypass; ONCAB: on-pump coronary artery bypass; PS: propensity score; SMD: standardised

437

mean difference; BMI: body mass index; CCS: Canadian cardiovascular class; NYHA: New York Heart Association; MI:

438

myocardial infarction; PCI: percutaneous coronary intervention; DM diabetes mellitus; sCr: serum creatinine; COPD: chronic

439

obstructive pulmonary disease; CVA: cerebrovascular accident; PVD: peripheral vascular disease; AF: atrial fibrillation; TVD: 3-

440

vessel disease; LMD: left main disease; LVEF: left ventricular ejection fraction; IABP: intra-aortic balloon pump.

ACCEPTED MANUSCRIPT

Table 2. Intraoperative data

Mean number of grafts/pt Graft target LAD

2.7±0.7

RI PT

P value

<0.001

SC

1 2 3 4 5 6

ONCAB n % 1197 100.0 6 0.5 253 21.1 703 58.7 222 18.5 12 1.0 1 0.1

M AN U

Overall Number of grafts

OPCAB n % 1253 100.0 36 2.9 447 35.7 650 51.9 119 9.5 1 0.1 0 0.0

3.0±0.7

35 1218

2.8 97.2

33 1164

2.8 97.2

1

DIA

No Yes

1020 233

81.4 18.6

931 266

77.8 22.2

0.03

CX

No Yes

313 940

25.0 75.0

142 1055

11.9 88.1

<0.001

RCA

No Yes

398 855

31.8 68.2

305 892

25.5 74.5

<0.001

RITA

No Yes

1224 29

97.7 2.3

1160 37

96.9 3.1

0.3

LITA

No Yes

76 1177

6.1 93.9

113 1084

9.4 90.6

0.002

BITA

No Yes

1230 23

98.2 1.8

1163 34

97.2 2.8

0.1

RA

No

986

78.7

1083

90.5

<0.001

EP

TE D

No Yes

AC C

441 442

ACCEPTED MANUSCRIPT

267

21.3

114

9.5

No Yes

137 1116

10.9 89.1

58 1139

4.8 95.2

No 1154 92.1 1150 96.1 <0.001 Yes 99 7.9 47 3.9 LAD: left anterior descending artery; DIA: diagonal branch; CX: circumflex artery; RCA: right coronary artery; LITA: left interior thoracic artery; RITA: right interior thoracic artery; BITA: bilateral internal thoracic artery; RA: radial artery; SV: saphenous vein

AC C

EP

TE D

M AN U

SC

sequential grafts 443 444

<0.001

RI PT

SV

Yes

ACCEPTED MANUSCRIPT

Table 3. Short term outcomes in the unmatched and matched OPCAB and ONCAB groups with relative effect size in the propensity

446

score-matched population. (OPCAB: off-pump coronary artery bypass; ONCAB: on-pump coronary artery bypass; PS: propensity

447

score; CVA: cerebrovascular event; RRT: renal replacement therapy; SWR: sternal wound reconstruction; IABP intra-aortic balloon

448

pump)

449

% 100.0 98.6 1.4

n 1197 1167 30

RRT

No Yes

1193 60

95.2 4.8

1144 53

IABP

No Yes

1221 32

97.4 2.6

SWR

No Yes

1240 13

99.0 1.0

Re-exploration

No Yes

1224 29

Any complication

No Yes

30-day mortality

No Yes

PS-matched ONCAB

PSM-adjusted estimate [95%CI]

adjusted P

0.49[0.25-0.99]

0.04

% 100.0 97.5 2.5

n 995 983 12

% 100.0 98.8 1.2

n 995 971 24

% 100.0 97.6 2.4

95.6 4.4

946 49

95.1 4.9

952 43

95.7 4.3

1.15[0.75-1.74]

0.52

1136 61

94.9 5.1

969 26

97.4 2.6

942 53

94.7 5.3

0.48[0.30-0.77]

0.002

1180 17

98.6 1.4

985 10

99.0 1.0

979 16

98.4 1.6

0.62[0.28-1.38]

0.24

97.7 2.3

1155 42

96.5 3.5

973 22

97.8 2.2

956 39

96.1 3.9

0.55[0.33-0.94]

0.02

1118 135

89.2 10.8

1028 169

85.9 14.1

890 105

89.4 10.6

849 146

85.3 14.7

0.69[0.52-0.90]

0.005

1221 32

97.4 2.6

1171 26

97.8 2.2

970 25

97.5 2.5

976 19

98.1 1.9

1.32[0.73-2.40]

0.36

AC C

EP

No Yes

n 1253 1235 18

Sample size CVA

PS-matched OPCAB

M AN U

unmatched ONCAB

TE D

unmatched OPCAB

SC

RI PT

445

ACCEPTED MANUSCRIPT

450

No 1062 84.8 1102 92.1 865 86.9 921 92.6 2[1.50-2.7] <0.001 Yes 191 15.2 95 7.9 139 14.0 74 7.4 OPCAB: off-pump coronary artery bypass; ONCAB: on-pump coronary artery bypass; PS: propensity score; CVA: cerebrovascular

451

event; RRT: renal replacement therapy; SWR: sternal wound reconstruction; IABP intra-aortic balloon pump

RI PT

IR

AC C

EP

TE D

M AN U

SC

452

ACCEPTED MANUSCRIPT

COPD

sCr

19.518

CCS

BMI

10.424

AF

LMD

7.494

DMI

MI30d

7.322

TVD

1.984

CVA

1.508

LVEF

1.27

Current smoking

0.701

NYHA

0.688

PVD

0.661

SC

19.776

M AN U

preoperative IABP

RI PT

Supplementary Table 1. Percentage of missing data

0.622 0.622 0.615

PCI

0.463

shock

0.417

non elective

0.093

emergent/salvage

0.093

TE D

EP

0.635

Age

0

Female

0

Era of surgery

0

AC C

453

454

BMI: body mass index; CCS: Canadian cardiovascular class; NYHA: New York Heart Association; MI: myocardial infarction; PCI:

455

percutaneous coronary intervention; DM diabetes mellitus; sCr: serum creatinine; COPD: chronic obstructive pulmonary disease;

456

CVA: cerebrovascular accident; PVD: peripheral vascular disease; AF: atrial fibrillation; TVD: 3-vessel disease; LMD: left main

457

disease; LVEF: left ventricular ejection fraction; IABP: intra-aortic balloon pump.

458

ACCEPTED MANUSCRIPT

460

ONCAB groups with relative effect size in the propensity score-matched population. unmatched ONCAB n % 706 100.0 685 97.0 21 3.0

RRT

No Yes

706 27

96.3 3.7

683 23

96.7 3.3

IABP

No Yes

714 19

97.4 2.6

672 34

95.2 4.8

SWR

No

728

99.3

696

Yes

5

0.7

10

No

717

97.8

Yes

16

2.2

Any No complication Yes

663

90.5

70

9.5

30-day mortality

No

717

Yes No

IR

M AN U

adjusted estimate [95%CI]

adjusted P

0.31[0.11-0.84]

0.02

562 21

96.4 3.6

565 19

96.7 3.3

1.11[0.592.09]

0.74

569 14

97.6 2.4

555 29

95.0 5.0

0.47[0.25-0.90]

0.02

0.44[0.14-1.44]

0.17

0.53[0.27-1.05]

0.07

0.63[0.43-0.90]

0.01

0.89[0.38-2.09]

0.78

2.2[1.51-3.3]

<0.001

579

99.3

575

98.5

1.4

4

0.7

9

1.5

681

96.5

570

97.8

560

95.9

25

3.5

13

2.2

24

4.1

614

87.0

531

91.1

505

86.5

92

13.0

52

8.9

79

13.5

97.8

691

97.9

573

98.3

573

98.1

16

2.2

15

2.1

10

1.7

11

1.9

621

84.7

653

92.5

499

85.6

543

93.0

EP

Reexploration

PS-matched ONCAB n % 584 100.0 568 97.3 16 2.7

98.6

AC C

Sample size CVA

PS-matched OPCAB n % 583 100.0 578 99.1 5 0.9

SC

No Yes

unmatched OPCAB n % 733 100.0 724 98.8 9 1.2

RI PT

Supplementary Table 2. Short term outcomes in patients with orally treated diabetes in the unmatched and matched OPCAB and

TE D

459

ACCEPTED MANUSCRIPT

462

event; RRT: renal replacement therapy; SWR: sternal wound reconstruction; IABP intra-aortic balloon pump

RI PT

461

Yes 112 15.3 53 7.5 84 14.4 41 7.0 OPCAB: off-pump coronary artery bypass; ONCAB: on-pump coronary artery bypass; PS: propensity score; CVA: cerebrovascular

463

AC C

EP

TE D

M AN U

SC

464

ACCEPTED MANUSCRIPT

Supplementary Table 3. Short term outcomes in patients on insulin in the unmatched and matched OPCAB and ONCAB groups

466

with relative effect size in the propensity score-matched population. (OPCAB: off-pump coronary artery bypass; ONCAB: on-pump

467

coronary artery bypass; PS: propensity score; CVA: cerebrovascular event; RRT: renal replacement therapy; SWR: sternal wound

468

reconstruction; IABP intra-aortic balloon pump)

% 100.0 98.3 1.7

n 491 482 9

% 100 98 2

RRT

No Yes

487 33

93.7 6.3

461 30

94 6

IABP

No Yes

507 13

97.5 2.5

464 27

SWR

No Yes

512 8

98.5 1.5

Reexploration

No

507

Yes

95 5

EP

Sample size CVA

TE D

No Yes

n 520 511 9

PS-matched OPCAB n 412 405 7

SC

unmatched ONCAB

PS-matched ONCAB

adjusted estimate [95%CI]

adjusted P

0.87[0.31-2.42]

0.79

% 100.0 98.3 1.7

n 411 403 8

% 100.0 98.1 1.9

384 28

93.2 6.8

387 24

94.2 5.8

1.18[0.67-2.06]

0.57

400 12

97.1 2.9

387 24

94.2 5.8

0.48[0.24-0.98]

0.04

M AN U

unmatched OPCAB

RI PT

465

99 1

406 6

98.5 1.5

404 7

98.3 1.7

0.85[0.28-2.56]

0.77

97.5

474

97

403

97.8

396

96.4

0.59[0.26-1.36]

0.21

13

2.5

17

3

9

2.2

15

3.6

Any No complication Yes

455

87.5

414

84

359

87.1

344

83.7

0.77[0.54-0.99]

0.04

65

12.5

77

16

53

12.9

67

16.3

30-day

504

96.9

480

98

397

96.4

403

98.1

1.81[0.77-4.28]

0.19

No

AC C

484 7

ACCEPTED MANUSCRIPT

mortality Yes

16

3.1

11

2

15

8

1.9

469

No 441 84.8 449 91 357 86.7 378 92.0 1.76[1.12-2.8] 0.01 Yes 79 15.2 42 9 55 13.3 33 8.0 OPCAB: off-pump coronary artery bypass; ONCAB: on-pump coronary artery bypass; PS: propensity score; CVA: cerebrovascular

470

event; RRT: renal replacement therapy; SWR: sternal wound reconstruction; IABP intra-aortic balloon pump

AC C

EP

TE D

M AN U

SC

RI PT

IR

3.6

ACCEPTED MANUSCRIPT

Supplementary Table 4. Effect of off-pump coronary artery bypass (OPCAB) grafting over on-pump coronary artery bypass

472

(ONCAB) grafting on outcomes of interest in a fully adjusted double robust analyisis.

RI PT

471

lower 95%CI

upper 95%CI

P

0.23

0.97

0.04

0.77

1.97

0.55

0.14

0.88

0.04

0.57*

0.33

0.99

0.04

0.49*

0.3

0.82

0.009

0.7*

0.51

0.95

0.02

1.4Ŧ

0.75

2.61

0.19

2.41*

1.75

3.32

<0.0001

1.09 Ŧ

0.91

1.29

0.55

0.47*

Postoperative RRT

1.23*

Sternal wound reconstruction

0.35*

M AN U

Postoperative CVA

Re-exploration for bleeding Postoperative IABP

Late mortality

EP

Incomplete revascularization

TE D

Any of above complication Mortality at 30 days

SC

Effect size

*Logistic regression model; Ŧ Cox regression model; CI: confidence interval

474

CVA: cerebrovascular event; RRT: renal replacement therapy; IABP: intra-aortic balloon pump.

475

476

477

AC C

473

ACCEPTED MANUSCRIPT

TE D

EP

0.47 1.25 1.17 1.43 0.7 1.29 0.81 1.26 1.03 1.44 2.3 0.33 4.16 2.1 0.65 1.37 1.61 0.62 0.49 4.5 2.75 1.57 2.76 1.81 2.21 0.8 1.01

AC C

OPCAB Age Female Body mass index Canadian Cardiovascular Class New Your Heart Association Class Myocardial Infarction within 30 day Percutaneous Coronary intervention Diabetes Mellitus on insulin Current smoking Creatinine > 200 mmol/l Chronic Obstructive Pulmonary Disease Cerebrovascular accident Peripheral vascular disease Atrial fibrillation 3-vessel disease Left main disease Left ventricular ejection fraction 30%-49% Left ventricular ejection fraction <30% Cardiogenic shock preop intraaortic balloon pump non elective emergent/salvage Era of surgery Bilateral internal thoracic arteries Radial artery Incomplete revascularization 479

Lower 0.95CI 0.23 0.71 0.51 0.82 0.32 0.59 0.33 0.39 0.5 0.5 0.62 0.07 1.67 0.94 0.08 0.57 0.75 0.27 0.1 0.26 0.47 0.71 0.4 1 0.26 0.23 0.33

Upper 0.95CI 0.97 2.18 2.67 2.51 1.5 2.8 1.97 4.01 2.11 4.16 8.48 1.52 10.36 4.68 5.02 3.32 3.43 1.46 2.53 76.93 16.09 3.49 18.91 3.28 18.59 2.83 3.06

M AN U

Effect

RI PT

Supplementary Table 5. Double robust analysis on postoperative cerebrovascular accident.

SC

478

ACCEPTED MANUSCRIPT

TE D

EP

1.23 1.22 0.76 1.27 1.25 0.73 0.66 0.74 1.91 0.98 5.24 0.66 2.11 1.47 1.92 0.89 1.44 0.94 1.26 4.26 3.68 1.05 1.16 1.95 0.94 0.76 0.75

AC C

OPCAB Age Female Body mass index Canadian Cardiovascular Class New Your Heart Association Class Myocardial Infarction within 30 day Percutaneous Coronary intervention Diabetes Mellitus on insulin Current smoking Creatinine > 200 mmol/l Chronic Obstructive Pulmonary Disease Cerebrovascular accident Peripheral vascular disease Atrial fibrillation 3-vessel disease Left main disease Left ventricular ejection fraction 30%-49% Left ventricular ejection fraction <30% Cardiogenic shock preop intraaortic balloon pump non elective emergent/salvage Era of surgery Bilateral internal thoracic arteries Radial artery Incomplete revascularization 481 482

Lower 0.95CI 0.77 0.86 0.43 0.92 0.78 0.45 0.36 0.32 1.22 0.45 2.65 0.31 1.04 0.85 0.82 0.53 0.88 0.56 0.56 0.57 1.08 0.64 0.18 1.29 0.12 0.36 0.35

Upper 0.95CI 1.97 1.74 1.34 1.74 2.01 1.21 1.2 1.72 3 2.11 10.35 1.41 4.27 2.53 4.49 1.48 2.37 1.57 2.86 31.57 12.56 1.72 7.38 2.94 7.34 1.61 1.59

M AN U

Effect

RI PT

Supplementary Table 6. Double robust analysis on postoperative renal replacement therapy.

SC

480

ACCEPTED MANUSCRIPT

483

TE D

Upper 0.95CI 0.82 1.43 2.24 1.11 2.12 2.25 3.1 1.97 1.96 0.84 6.09 1.66 2.24 1.55 4.66 1.65 2.29 3.76 15.93 13.66 2.4 1.57 19.29 2.66 14.33 1.2 2.03

SC

0.49 0.97 1.27 0.82 1.22 1.29 1.68 0.78 1.18 0.24 2.5 0.65 0.3 0.77 1.74 0.91 1.31 2.09 8.3 1.44 0.14 0.87 6.34 1.75 3.58 0.36 0.92

EP

OPCAB Age Female Body mass index Canadian Cardiovascular Class New Your Heart Association Class Myocardial Infarction within 30 day Percutaneous Coronary intervention Diabetes Mellitus on insulin Current smoking Creatinine > 200 mmol/l Chronic Obstructive Pulmonary Disease Cerebrovascular accident Peripheral vascular disease Atrial fibrillation 3-vessel disease Left main disease Left ventricular ejection fraction 30%-49% Left ventricular ejection fraction <30% Cardiogenic shock preop intraaortic balloon pump non elective emergent/salvage Era of surgery Bilateral internal thoracic arteries Radial artery Incomplete revascularization

Lower 0.95CI 0.3 0.65 0.72 0.6 0.7 0.74 0.91 0.31 0.71 0.07 1.03 0.26 0.04 0.38 0.65 0.5 0.75 1.17 4.32 0.15 0.01 0.48 2.08 1.15 0.89 0.11 0.42

M AN U

Effect

RI PT

Supplementary Table 7. Double robust analysis on need for intra-aortic balloon pump.

AC C

484

ACCEPTED MANUSCRIPT

TE D

EP

0.35 2.52 0.52 2.23 1.07 0.39 0.43 2.06 1.53 0.81 4.56 3.62 0.68 0.61 1.25 0.33 1.12 2.08 2.95 1.3 2 2.07 1.01

AC C

OPCAB Age Female Body mass index Canadian Cardiovascular Class New Your Heart Association Class Myocardial Infarction within 30 day Percutaneous Coronary intervention Diabetes Mellitus on insulin Current smoking Creatinine > 200 mmol/l Chronic Obstructive Pulmonary Disease Cerebrovascular accident Peripheral vascular disease Atrial fibrillation 3-vessel disease Left main disease Left ventricular ejection fraction 30%-49% Left ventricular ejection fraction <30% Cardiogenic shock preop intraaortic balloon pump non elective emergent/salvage Era of surgery Bilateral internal thoracic arteries Radial artery Incomplete revascularization 486

Lower 0.95CI 0.14 1.19 0.15 0.97 0.44 0.15 0.13 0.67 0.65 0.16 1.26 1.4 0.08 0.17 0.27 0.13 0.44 0.84 0.75 0.51 0.7 0.62 0.21

Upper 0.95CI 0.88 5.34 1.84 5.11 2.59 1.04 1.43 6.33 3.6 3.97 16.52 9.39 5.56 2.18 5.8 1.01 2.87 5.13 11.67 3.34 5.72 6.93 4.92

M AN U

Effect

RI PT

Supplementary Table 8. Double robust analysis on risk of sternal wound reconstruction.

SC

485

ACCEPTED MANUSCRIPT

TE D

EP

0.57 0.98 1.1 0.91 1.35 0.55 1.37 0.33 0.93 0.68 2.77 1.39 0.54 0.72 2.18 1.11 0.7 0.72 1.66 13.13 2.54 4.85 1.08 1.81 0.36 0.85

AC C

OPCAB Age Female Body mass index Canadian Cardiovascular Class New Your Heart Association Class Myocardial Infarction within 30 day Percutaneous Coronary intervention Diabetes Mellitus on insulin Current smoking Creatinine > 200 mmol/l Chronic Obstructive Pulmonary Disease Cerebrovascular accident Peripheral vascular disease Atrial fibrillation 3-vessel disease Left main disease Left ventricular ejection fraction 30%-49% Left ventricular ejection fraction <30% Cardiogenic shock preop intraaortic balloon pump non elective emergent/salvage Era of surgery Bilateral internal thoracic arteries Radial artery Incomplete revascularization 488 489

Lower 0.95CI 0.33 0.64 0.59 0.64 0.74 0.29 0.71 0.08 0.53 0.25 1.03 0.59 0.13 0.32 0.8 0.58 0.37 0.37 0.71 1.66 1.31 1.21 0.67 0.37 0.11 0.33

Upper 0.95CI 0.99 1.5 2.06 1.31 2.44 1.03 2.65 1.44 1.64 1.84 7.47 3.29 2.3 1.65 5.94 2.13 1.33 1.4 3.9 104.09 4.93 19.39 1.75 8.83 1.2 2.14

M AN U

Effect

RI PT

Supplementary Table 9. Double robust analysis on risk of re-exploration for bleeding.

SC

487

ACCEPTED MANUSCRIPT

Supplementary Table 10. Double robust analysis on incidence of any complication among cerebrovascular accident, renal

491

replacement therapy, need for intraaortic balloon pump, sternal wound reconstruction and re-exploration for bleeding.

TE D

EP

492

Upper 0.95CI 0.95 1.05 1.20 1.06 1.96 1.50 1.19 1.54 1.85 1.31 8.75 1.38 2.81 1.64 2.34 1.08 2.06 1.83 4.90 4.38 3.88 1.52 11.81 2.41 2.37 1.18 1.71

SC

0.7 1.03 0.82 1.03 1.43 1.08 0.84 1.06 1.11 0.77 5.15 0.85 1.65 1.11 1.27 0.75 1.47 1.3 3.04 0.85 1.43 1.07 4.72 1.65 1.27 0.71 1.05

AC C

OPCAB Age Female Body mass index Diabetes mellitus on insulin Canadian Cardiovascular Class New Your Heart Association Class Myocardial Infarction within 30 day Percutaneous Coronary intervention Current smoking Creatinine > 200 mmol/l Chronic Obstructive Pulmonary Disease Cerebrovascular accident Peripheral vascular disease Atrial fibrillation 3-vessel disease Left main disease Left ventricular ejection fraction 30%-49% Left ventricular ejection fraction <30% Cardiogenic shock preop intraaortic balloon pump non elective emergent/salvage Era of surgery Bilateral internal thoracic arteries Radial artery Incomplete revascularization

Lower 0.95CI 0.51 1.01 0.56 1 1.05 0.77 0.6 0.72 0.64 0.42 2.98 0.5 0.92 0.73 0.63 0.52 1.05 0.91 1.85 0.13 0.44 0.75 1.8 1.14 0.72 0.41 0.62

M AN U

Effect

RI PT

490

ACCEPTED MANUSCRIPT

TE D

EP

1.4 1.05 1.64 0.81 1.65 1.17 1.01 1.49 1.49 0.24 1.03 0.59 1.5 1.54 2.16 1.39 0.83 2.59 3.79 1.24 0.54 2.25 1.55 0.76 4.01 0.35 0.79

AC C

OPCAB Age Female Body mass index Canadian Cardiovascular Class New Your Heart Association Class Myocardial Infarction within 30 day Percutaneous Coronary intervention Diabetes Mellitus on insulin Current smoking Creatinine > 200 mmol/l Chronic Obstructive Pulmonary Disease Cerebrovascular accident Peripheral vascular disease Atrial fibrillation 3-vessel disease Left main disease Left ventricular ejection fraction 30%-49% Left ventricular ejection fraction <30% Cardiogenic shock preop intraaortic balloon pump non elective emergent/salvage Era of surgery Bilateral internal thoracic arteries Radial artery Incomplete revascularization 494 495

Lower 0.95CI 0.75 1.005 0.86 0.55 0.81 0.59 0.48 0.49 0.79 0.03 0.23 0.17 0.52 0.74 0.83 0.62 0.42 1.3 1.54 0.06 0.03 1.04 0.27 0.4 0.49 0.08 0.3

Upper 0.95CI 2.61 1.09 3.14 1.2 3.36 2.29 2.13 4.54 2.8 1.78 4.54 1.99 4.32 3.18 5.66 3.14 1.66 5.17 9.32 25.92 10.61 4.86 9 1.46 33.02 1.51 2.11

M AN U

Effect

RI PT

Supplementary Table 11. Double robust analysis on mortality within 30 days

SC

493

ACCEPTED MANUSCRIPT

496 497

Supplementary Table 12. Double robust analysis on incomplete revascularization.

499 500

SC

Upper 0.95 3.32 1.49 1.96 1.14 1.2 1.65 1.19 1.21 1.35 1.46 2.17 1.86 1.87 1.53 1.91 1.71 1.67 1.77 5.47 3.68 1.64 4.54 1.93 4.41 0.35 0.91

TE D

M AN U

Lower 0.95 1.75 0.92 1 0.78 0.63 0.85 0.53 0.35 0.73 0.52 0.52 0.72 0.55 0.67 0.48 0.86 0.85 0.55 0.05 0.35 0.84 0.43 1.13 0.5 0.13 0.33

EP

Effect 2.41 1.17 1.4 0.94 0.87 1.18 0.79 0.66 0.99 0.87 1.07 1.16 1.01 1.01 0.95 1.21 1.19 0.99 0.52 1.14 1.18 1.4 1.48 1.48 0.21 0.56

AC C

OPCAB Age Female Body mass index Canadian Cardiovascular Class New Your Heart Association Class Myocardial Infarction within 30 day Percutaneous Coronary intervention Diabetes Mellitus on insulin Current smoking Creatinine > 200 mmol/l Chronic Obstructive Pulmonary Disease Cerebrovascular accident Peripheral vascular disease Atrial fibrillation 3-vessel disease Left main disease Left ventricular ejection fraction 30%-49% Left ventricular ejection fraction <30% Cardiogenic shock preop intraaortic balloon pump non elective emergent/salvage Era of surgery Bilateral internal thoracic arteries Radial artery

RI PT

498

ACCEPTED MANUSCRIPT

Supplementary Table 13. Double robust analysis on late mortality.

502 503 504 505 506

SC

RI PT

Upper 0.95 1.29 1.07 1.12 1.27 1.23 1.28 0.95 1.59 1.4 1.72 3.72 1.83 2.42 2.14 2.52 1.07 1.35 1.59 3.03 2.44 3.7 1.42 2.41 1.23 1.77 1.36 1.92

M AN U

Lower 0.95 0.91 1.04 0.76 1.01 0.84 0.87 0.54 0.66 1 0.93 1.68 1.03 1.34 1.39 1.32 0.71 0.9 1.08 1.71 0.09 0.39 0.97 0.53 0.66 0.43 0.84 1.15

TE D

AC C

OPCAB Age Female Body mass index Canadian Cardiovascular Class New Your Heart Association Class Myocardial Infarction within 30 day Percutaneous Coronary intervention Diabetes Mellitus on insulin Current smoking Creatinine > 200 mmol/l Chronic Obstructive Pulmonary Disease Cerebrovascular accident Peripheral vascular disease Atrial fibrillation 3-vessel disease Left main disease Left ventricular ejection fraction 30%-49% Left ventricular ejection fraction <30% Cardiogenic shock preop intraaortic balloon pump non elective emergent/salvage Era of surgery Bilateral internal thoracic arteries Radial artery Incomplete revascularization

Effect 1.09 1.06 0.92 1.13 1.02 1.06 0.72 1.02 1.18 1.27 2.5 1.38 1.8 1.72 1.82 0.87 1.1 1.31 2.28 0.48 1.2 1.18 1.13 0.91 0.87 1.07 1.49

EP

501

emergent/ salvage

BMI

Era of Surgery

TVD

IR

OPCAB

0.10

0.09

0.04

0.04

0.00

0.09

-0.02

-0.14

0.16

0.09

0.05

0.01

0.01

-0.02

0.00

-0.05

0.04

-0.01

0.06

0.01

0.06

-0.09

-0.04

0.05

-0.00

-0.01

0.02

-0.01

0.03

0.03

0.03

0.03

0.23

0.11

0.08

-0.15

-0.01

-0.00

-0.02

0.15

0.02

0.08

0.04

-0.01

0.12

0.05

0.03

0.07

0.12

0.13

-0.06

0.02

0.03

-0.01

0.02

-0.01

-0.01

0.00

0.10

0.11

0.08

0.05

0.38

0.11

0.01

0.30

0.06

-0.00

-0.03

0.04

-0.02

-0.02

-0.02

0.01

0.01

0.03

0.06

0.04

0.08

0.03

0.19

-0.06

-0.02

0.02

CVA

PVD

AF

LMD

LVEF

RI PT

non elective

0.03

-0.02 0.01

SC

IABP

0.04

-0.04

M AN U

Cardiogenic shock

0.07

0.05

TE D

Pearson correlation between predictors VI F

COPD

0.04

-0.00

0.04

-0.02

0.00

sCr>200mmol/l

1.0 6

0.03

0.06 0.03

-0.01

PCI

0.03

0.14

0.00

0.03

1.00

1.3 4

0.04

-0.01

-0.24 0.08

0.04

0.34

MI

0.03

-0.02

Current smoking

0.03

0.34

1.2 4

0.02

0.03

-0.16

0.02

1.00

NYHA

0.12

0.02

-0.02

0.12

0.15

0.08

DM on insulin

0.01

0.15

1.2 5

-0.02

1.00

PCI

0.09

1.00

CCS

0.01

0.07

MI

-0.00

1.0 6

0.09

0.03

NYHA

0.01

Female

-0.00

0.07

CCS

1.2 4

0.01

1.00

Female

Age

1.00

-0.01

Age

EP

Supplementary Table 14. Person correlation coefficient and variance inflation factors (VIF) from the covariance matrix 507

AC C

ACCEPTED MANUSCRIPT

-0.05

0.01

-0.00

0.00

0.06

-0.03

0.01

0.07

0.00

-0.01

-0.00

-0.02

0.01

-0.02

-0.03

0.05

0.04

0.02

0.01

-0.02

0.01

0.04

-0.02

0.06

0.14

0.04

0.03

-0.01

-0.01

-0.02

0.01

0.02

-0.01

-0.00

-0.00

0.05

0.00

0.02

0.03

0.06

-0.03

-0.04

0.03

-0.01

-0.01

0.04

0.01

0.01

-0.01

0.01

0.09

0.02

-0.03

0.03

-0.03

0.03

-0.01

-0.02

0.00

0.00

1.00

0.04

0.03

0.01

0.23

0.06

0.00

0.09

0.06

0.03

-0.02

0.04

1.00

0.09

0.07

0.13

0.08

-0.03

-0.01

0.09

0.03

0.00

-0.02 0.07

RI PT

0.09

0.00 0.02 0.04 -0.02

SC

-0.01

-0.02 0.05 0.04 0.01

0.09

Current smoking

1.1 2

sCr>200 mmol/l

1.0 3

COPD

1.0 7

CVA

1.0 2

PVD

1.0 3

AF

1.0 3

LMD

1.0 8

LVEF

1.0 8

TE D

1.00

1.0 7

EP

0.01

M AN U

0.03

0.02 -0.02

0.00

-0.01

-0.04 0.03

-0.03

0.03

-0.02

0.06

-0.02

0.03

0.05

0.01

-0.02

0.02

1.00

0.04

-0.03

-0.01

0.07

0.08

0.03

0.04

-0.02

-0.04

0.01

0.07

0.07 0.01

0.01 0.03

-0.02

0.01

0.11

-0.01

0.02 0.08 0.08 0.02

0.07

-0.02

0.10

0.12

-0.02

0.04 0.08 1.00 0.08 0.04

-0.02

0.00

-0.01

0.03

0.03

0.02 0.02 0.02 0.02 -0.02

-0.01

0.04

0.03

-0.05

0.08

0.09 1.00 0.09 0.04

-0.01

0.08

-0.01

0.00

0.04

1.00

-0.00 -0.00 0.05 0.02

0.02

0.02

-0.02

0.09

0.03

1.00 0.06 0.04 0.15

-0.01

0.01

0.10

0.04

0.03 0.14 0.04 0.01

-0.02

0.03

0.04

0.02 -0.01 0.00

-0.04

0.04

0.03

0.08 -0.02 -0.00

0.07

0.07

0.03 0.00 -0.02

0.03

0.08 -0.24

1.00

-0.16

DM on insulin

AC C

ACCEPTED MANUSCRIPT

-0.00

-0.01

-0.00

0.00

0.03

0.17

-0.00

0.11

0.01

0.01

-0.00

0.03

1.00

0.12

-0.03

0.06

0.08

0.02

-0.04

0.29

0.17

0.12

1.00

0.00

0.00

0.01

0.01

-0.01

-0.03

-0.04

-0.00

-0.03

0.00

1.00

0.15

-0.02

0.01

-0.01

0.09

-0.01

-0.00

0.11

0.06

0.00

0.15

1.00

0.02

0.06

0.06

0.06

0.09

-0.01

0.01

0.08

0.01

-0.02

0.02

1.00

0.21

-0.08

0.03

0.03

-0.00

0.01

0.02

0.01

0.01

0.06

0.21

1.00

0.11

-0.02

0.00

0.00

-0.00

-0.04

-0.01

-0.01

0.06

-0.08

0.11

1.00

1.00 0.09

RI PT

-0.04

1.00 0.04 0.08 0.00

SC

0.29

0.29 0.13 0.06

M AN U

0.04

0.07 0.23

0.29

0.01

0.03

EP

TE D

0.01

0.00

0.02

-0.00

0.00

-0.01

0.00

-0.02

0.04

-0.00

0.02

-0.01

0.03

0.00

-0.02

1.0 3

-0.03

0.03

-0.03

0.05

0.01

-0.00

OPCAB

-0.01

-0.03

0.02

0.07

-0.06

0.03

-0.01

-0.03

-0.05

0.06

1.0 7

-0.02

0.01

0.03

0.01

0.19

0.02

IR

-0.00

0.01

0.02

-0.02

0.09

0.30

-0.01

0.04

-0.04

0.04

-0.03

0.03

-0.06

1.0 9

-0.00

-0.01

0.01

0.01

-0.01

0.01

-0.15

TVD

0.05

-0.01

-0.03

0.01

0.06

0.08

0.13

1.3 4

-0.04

0.02

-0.02

-0.02

0.03

0.11

0.08

Era of surgery

-0.09

0.00

-0.02

0.00

0.04

0.12

0.06

0.05

0.02

0.00

0.38

0.11

1.1 1

0.01

-0.00

-0.02

0.06

0.07

0.01

BMI

0.05

-0.00

-0.02

0.05

0.23

1.1 5

0.09

-0.01

0.03

0.03

0.06

emergent/salva ge

0.16

-0.01

0.08

0.03

1.3 3

-0.14

0.03

0.05

-0.01

non elective

-0.02

0.04

0.03

1.1 3

0.09

0.14

0.04

IABP

0.00

0.06

1.1 9

0.04

-0.02

Cardiogenic Shock

AC C

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

508 OPCAB: off-pump coronary artery bypass; BMI: body mass index; CCS: Canadian cardiovascular class; NYHA: New York Heart

510

Association; MI: myocardial infarction; PCI: percutaneous coronary intervention; DM diabetes mellitus; sCr: serum creatinine;

511

COPD: chronic obstructive pulmonary disease; CVA: cerebrovascular accident; PVD: peripheral vascular disease; AF: atrial

512

fibrillation; TVD: 3-vessel disease; LMD: left main disease; LVEF: left ventricular ejection fraction; IABP: intra-aortic balloon

513

pump.

M AN U

SC

RI PT

509

AC C

EP

TE D

514

ACCEPTED MANUSCRIPT

Figure Legend

516

Figure 1. Number of off-pump coronary artery bypass (OPCAB) and on-pump coronary artery bypass (ONCAB) operations during

517

the study period

518

Figure 2. Survival curves in the unmatched and matched off-pump coronary artery bypass (OPCAB) and on-pump coronary artery

519

bypass (ONCAB) groups.

520

Figure 3. Survival curves in the matched off-pump coronary artery bypass (OPCAB) and on-pump coronary artery bypass (ONCAB)

521

groups in patients with oral treatment and in patients on insulin

522

Figure 4. Survival curves in the matched off-pump coronary artery bypass (OPCAB) and on-pump coronary artery bypass (ONCAB)

523

groups in patients with complete (CR) and incomplete revascularization (IR)

524

Central picture. Effect of off-pump coronary artery bypass (OPCAB) grafting over on-pump coronary artery bypass (ONCAB)

525

grafting on outcomes of interest. (CVA: cerebrovascular event; RRT: renal replacement therapy; SW: sternal wound; IABP intra-

526

aortic balloon pump)

527

Supplementary Figure 1. Area under the curve (AUC) for propensity score.

528

Video 1. Off-pump coronary artery bypass grafting technique at Bristol Heart Institute.

AC C

EP

TE D

M AN U

SC

RI PT

515

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT