Abstract

Objective: Long-term outcome after coronary artery bypass grafting is worse in diabetic than in non-diabetic patients. No data are currently available regarding survival rates of diabetic and non-diabetic patients after coronary revascularisation compared with cohorts from the general population in the Netherlands, which were matched for age and sex (normal Dutch survival). Methods: We retrospectively analysed the data from 10 626 patients who underwent coronary artery bypass grafting between January 1998 and December 2007. Of these, 8287 patients were non-diabetic, 1587 were non-insulin-dependent and 630 were insulin-dependent diabetic patients (122 patients were lost to follow-up). Survival of these patient groups was compared with the normal Dutch survival. Results: Multivariate analyses revealed non-insulin-dependent diabetes to be a risk factor for early mortality and both insulin-dependent and non-insulin-dependent diabetes as risk factors for late mortality. The 1-, 5- and 10-year survival rates for non-diabetic patients were 94.1% ± 0.3%, 86.8% ± 0.4% and 75.1% ± 1.7%, respectively, which was better than the normal Dutch survival. For insulin-dependent diabetic patients, 1-, 5- and 10-year survival rates were 90.3% ± 1.2%, 78.0% ± 2.0% and 60.5% ± 4.6%, respectively, and for non-insulin-dependent diabetic patients 91.4% ± 0.7%, 79.0% ± 1.3% and 58.9% ± 3.4%, respectively, which was worse than the normal Dutch survival. Conclusions: Non-insulin-dependent diabetes was a risk factor for early mortality and both types of diabetes were risk factors for late mortality after revascularisation. Compared with age- and sex-matched cohorts from the general Dutch population, the 10-year survival of non-diabetic patients was better; whereas the survival of both types of diabetic patients was worse.

Introduction

Diabetes is a risk factor for poor outcomes after percutaneous [1–3] and surgical myocardial revascularisation [4–12]. In patients undergoing coronary artery bypass grafting (CABG), the incidence of diabetes ranges from 12% to 38% [13–17]. Reports addressing the long-term mortality of diabetic patients after CABG vary, with an incidence ranging from 1.3% to 6.5% per patient-year [14–19]. In diabetic patients, revascularisation appears to provide better long-term survival than percutaneous interventions do [17–19]. In some studies, diabetes also has been described as a predictor of early mortality [20–22], but others could not confirm this finding [14,23]. However, little is known about the long-term post-CABG survival of patients with and without diabetes compared with the survival of the general population. In this study, we evaluated the effect of diabetes on early and late mortality after CABG. We also compared the long-term survival of diabetic and non-diabetic patients with the survival of cohorts of the general population in the Netherlands matched for age and sex (normal Dutch survival).

Patients and methods

Patients

In this study, data of all patients undergoing an isolated CABG at a single centre (Catharina Hospital, Eindhoven, the Netherlands) between January 1998 and December 2007 were analysed. Starting in January 1998, clinical data, including demographic data, risk factors and complications, were prospectively collected in a database. Approval of this study was obtained from the institution’s research review board. We defined type 1 diabetes as insulin-dependent diabetes and type 2 diabetes as non-insulin-dependent diabetes, which included patients treated with oral medication or nutritional therapy.

Survival of age- and sex-matched cohorts from the general population

For calculating survival of general population cohorts, data from the database of the Dutch Central Bureau for Statistics (CBS), which can be downloaded online (www.cbs.nl), were used. For each year of the study period, general population cohorts were matched with the patient groups for age and gender. The year of operation of the patients group is the starting point for the general population cohorts.

Operative techniques

All patients received short-acting anaesthetic drugs to facilitate early extubation. Extracorporeal circulation was performed using normothermic non-pulsatile flow. Cold crystalloid cardioplegia (‘St Thomas’ solution) or warm blood cardioplegia was used to induce and maintain cardioplegic arrest, according to the surgeon’s preference. All patients undergoing CABG with the use of extracorporeal circulation received a low dose of aprotinin (2 million Kallikrein Inactivating Unit) administered in the prime solution of the extracorporeal circulation. Patients undergoing off-pump surgery did not receive aprotinin.

Follow-up

Follow-up data concerning mortality were gathered using databases of health insurance companies. The data of 9% of the patients in the total patient group could not initially be retrieved from these databases. We therefore contacted the general practitioners by telephone to obtain information about the mortality data of those patients. For the remaining patients, we contacted the city authorities where the patients lived at the time of the operation. Early mortality was defined as any-cause mortality within 30 days postoperatively, whereas late mortality was defined as any-cause mortality that occurred later than 30 days after CABG.

Statistical analyses

Discrete variables were compared with the chi-square test and are presented as numbers and percentages. Continuous variables were compared with a Student’s t-test and the analysis of variance and are presented as means ± standard deviations. Univariate and multivariate logistic regression analyses were performed to investigate the effect of biomedical variables on early mortality. Multivariate analyses were used to test for the potentially confounding effect of biomedical and demographic factors on outcomes. Cox proportional hazard regression analyses were performed for the same analyses of late mortality. Values for P less than 0.05 were considered statistically significant. If significant at P ≪ 0.05, confounders were included into the multivariable logistic and Cox regression analyses. Long-term survival was described using the Kaplan–Meier method. Comparisons of long-term survival were done using log-rank statistics. The zero time point indicates the time of CABG. Hazard ratios (HRs) with 95% confidence intervals (CIs) are reported. All statistical analyses were performed using SPSS software (Statistical Product and Services Solutions, version 15.0, SPSS Inc., Chicago, IL, USA).

Results

During a 10-year period (January 1998 and December 2007) 10 626 patients underwent CABG in our hospital. Excluding 122 patients who were lost to follow-up, 8287 patients (78.9%) had no diabetes, 1587 of the patients (15.1%) had type 2 diabetes and 630 patients (6%) had type 1 diabetes. Almost all patients who were lost to follow-up were non-Dutch or were living abroad. The minimum follow-up for surviving patients was 60 days. Mean follow-up was 1696 ± 1026 days (range, 0–3708 days) (0 day for operative deaths).

Baseline characteristics stratified by diabetes type are shown in Table 1 . Patients with type 2 diabetes were older and patients with type 1 diabetes were more often female. Impaired renal function, defined as a creatinine clearance level (CrCl) less than 60 ml min−1 per 1.73 m2, hypertension, body mass index greater than 35 kg m−2 and peripheral vascular disease (PVD) were more common in patients with diabetes (both types) than in non-diabetic patients.

Table 1

Baseline characteristics stratified by diabetes type.

Table 1

Baseline characteristics stratified by diabetes type.

Table 2 shows the incidences of early and late mortality stratified by type of diabetes. Early and late mortality are more frequent in both type 1 and type 2 diabetic patients. The annual incidence of death was 2.5 per 100 patient-years among non-diabetic patients, 4.1 per 100 patients-years among type 2 diabetic patients and 4.5 per 100 patient-years among type 1 diabetic patients. During the study period, the overall early mortality rate in our department fell from 2.9% in 1998 to 1% in 2007.

Table 2

Early and late mortality stratified by diabetes type.

Table 2

Early and late mortality stratified by diabetes type.

Risk factors for early mortality identified by univariate and multivariate logistic regression analyses are shown in Table 3 . Univariate logistic regression analysis revealed type 2 diabetes but not type 1 diabetes as a risk factor for early mortality. Other risk factors were age, chronic obstructive pulmonary disease (COPD), low CrCl, ejection factor (EF) less than 35%, PVD, previous cardiac surgery and emergency operation. Sex, hypertension, body mass index >35 kg m−2 and use of extracorporeal circulation were not identified as risk factors for early mortality. Complications such as perioperative myocardial infarction, perioperative use of intra-aortic balloon pump (IABP) and re-exploration also were identified as risk factors for early mortality.

Table 3

Predictors of early mortality; univariate and multivariate logistic regression analyses.

Table 3

Predictors of early mortality; univariate and multivariate logistic regression analyses.

All preoperative risk factors that were identified with the univariate logistic regression analyses were entered into the multivariate logistic regression model. Type 2 diabetes was an independent risk factor for early mortality. Other risk factors were age, COPD, low CrCl, EF less than 35%, previous cardiac surgery and emergency operation. PVD did not prove to be a risk factor.

Results of Cox regression analyses for risk factors of late mortality are shown in Table 4 . Univariate analyses revealed both types 1 and 2 diabetes as risk factors for late mortality. Other patient- and operation-related risk factors were age, COPD, low CrCl, EF less than 35%, PVD, previous cardiac surgery, female sex, emergency operation and hypertension. Furthermore, complications such as perioperative myocardial infarction, perioperative use of IABP and re-exploration were identified as risk factors for late mortality.

Table 4

Predictors of late mortality; univariate and multivariate Cox regression analyses.

Table 4

Predictors of late mortality; univariate and multivariate Cox regression analyses.

All preoperative risk factors identified with univariate analyses were entered into the multivariate Cox regression model. Both type 1 and type 2 diabetes were identified as independent risk factors for late mortality. Other risk factors were age, COPD, low CrCl, EF less than 35%, PVD, previous cardiac surgery and male sex. However, hypertension, emergency operation and use of extracorporeal circulation were not identified as independent risk factors.

Figs. 1–3 shows the long-term survival rates stratified by diabetes type and the corresponding normal Dutch survival. Non-diabetic patients had a better survival than type 1 and 2 diabetic patients (P ≪ 0.0001) (Fig. 1). Differences between type 1 and 2 diabetic patients were not significant (P = 0.483).

Fig. 1

Kaplan–Meier survival curves of non-diabetic, insulin-dependent and non-insulin-dependent diabetic patients. Log-rank diabetic versus non-diabetic: P value ≪0.0001. Log-rank type 1 versus type 2 diabetes: P value = 0.483.

Fig. 1

Kaplan–Meier survival curves of non-diabetic, insulin-dependent and non-insulin-dependent diabetic patients. Log-rank diabetic versus non-diabetic: P value ≪0.0001. Log-rank type 1 versus type 2 diabetes: P value = 0.483.

Fig. 2

Kaplan–Meier survival curves of non-diabetic patients and the survival of the normal Dutch population. Log-rank non-diabetic versus population: P value ≪0.0001.

Fig. 2

Kaplan–Meier survival curves of non-diabetic patients and the survival of the normal Dutch population. Log-rank non-diabetic versus population: P value ≪0.0001.

Fig. 3

Kaplan–Meier survival curves of insulin-dependent and non-insulin-dependent diabetic patients and the survival of the normal Dutch population. Log-rank typte-1 diabetes versus population: P value ≪0.0001. Log-rank type-2 diabetes versus population: P value ≪0.0001.

Fig. 3

Kaplan–Meier survival curves of insulin-dependent and non-insulin-dependent diabetic patients and the survival of the normal Dutch population. Log-rank typte-1 diabetes versus population: P value ≪0.0001. Log-rank type-2 diabetes versus population: P value ≪0.0001.

Non-diabetic patients had a better survival than the normal Dutch survival (P ≪ 0.0001) (Fig. 2). For non-diabetic patients, the 1-, 5- and 10-year survival rates were 94.1% ± 0.3%, 86.8% ± 0.4% and 75.1% ± 1.7%, respectively. For type 1 and type 2 diabetes patients, survival was worse than the normal Dutch survival (P ≪ 0.0001) (Fig. 3). For type 1 diabetic patients, the 1-, 5- and 10-year survival rates were 90.3% ± 1.2%, 78.0% ± 2.0% and 60.5% ± 4.6%, respectively, and for type 2 diabetic patients 91.4% ± 0.7%, 79.0% ± 1.3% and 58.9% ± 3.4%, respectively.

Discussion

Both insulin-dependent (type 1) and non-insulin-dependent diabetes (type 2) patients had worse long-term survival rates after a CABG than did their non-diabetic counterparts. When compared with age- and sex-matched cohorts from the general Dutch population, non-diabetic patients had better outcomes; whereas patients with either type 1 or type 2 diabetes had worse outcomes. Differences were larger between the survival rates of type 1 diabetic patients and the normal Dutch survival rates than between the survival rates of type 2 diabetic patients and the normal Dutch survival rates.

Criteria for defining diabetes vary in several studies. Diabetes was defined as the need for diet, oral medication or insulin in one study [14], while another study defined diabetes as the need for oral medication or insulin whereas nutritional therapy was not included. This might explain the differences in outcomes in several studies. In our study, we defined type 1 diabetes as insulin-dependent diabetes and type 2 diabetes as diabetes requiring oral or nutritional therapy. Twenty-one percent of our patient population was diabetic while other studies reported a prevalence of patients with diabetes ranging from 12% to 40% [12–17,20].

Early mortality

In this study, early mortality was 2.1% for non-diabetic patients, 3.2% for type 2 diabetes patients and 3.0% for type 1 diabetes patients (P = 0.013). Throughout the study, the overall early mortality rate decreased from 2.9% in 1998 to 1% in 2007. Several other studies have confirmed the higher incidence of early mortality in diabetic patients [14,19,21–23]. In our study, multivariate regression analysis identified only type 2 diabetes as a risk factor for early mortality. Type 1 diabetes did not prove to be an independent risk factor. It is plausible that type 2 diabetes patients were not well controlled preoperatively (and should have been on insulin), leading to a higher early mortality. Other studies provided conflicting information about the effect of diabetes on early mortality [11,14,24]. In our series, the odds ratio for type 2 diabetes was significant but rather low (1.49) compared with the odds ratios of other risk factors. Type 1 diabetes did not even prove to be a risk factor, indicating that only a small increase in risk for early mortality in diabetic patients can be expected. Although type 1 diabetes patients were relatively younger, the effect of age as a confounder is eliminated by using the logistic regression analysis.

Late mortality

In this study, the incidence of late mortality was higher in type 1 and type 2 diabetes patients (P ≪ 0.0001) than it was in non-diabetic patients. The incidence of death was 2.5 per 100 patient-years for non-diabetic patients, 4.1 per 100 patient-years for type 2 diabetes patients and 4.5 per 100 patient-years for type 1 diabetes patients. These findings are similar to those of several other studies [14,17–20]. The finding that type 1 diabetes affects only late mortality and not early mortality is possibly due to the long-term pathological effect of diabetes on different organic function. Deterioration of renal and neurological functions can be enhanced by the pathological effect of diabetes.

Taking into consideration that mortality rates will increase with increasing age, results will depend on the duration of the follow-up. With longer follow-up, higher numbers of deaths per 100 patient-years will be found. Thus, when follow-up is long, the 1- and 5-year survival rates will be more meaningful. The 1- and 5-year survival rates of non-diabetic as well as of type 1 and type 2 diabetic patients in this study were comparable with those found by others [14,21,24]. In contrast, Calafiore et al. [11] have shown that diabetes is an independent risk factor for early cardiac death only and the long-term survival is not statistically significantly different for diabetic and non-diabetic patients. However, their series consisted of relatively small number of patients (n = 2593) than ours (n = 10 514). Lack of statistical significance might simply be a lack of statistical power because of the reduced sample size [11].

Obviously, advanced age is associated with a poorer late survival rate. Type 1 diabetic patients, as shown in this study, are often younger. Although proper statistical testing can be used to distinguish between effects caused by age and diabetes, interpretation of survival curves remains difficult. Life expectancy may also vary with sex. In addition, variation in life expectancy over the years has been well documented in the Netherlands by the Central Bureau for Statistics. This organisation keeps track of mortality rates of the overall Dutch population. We used the Central Bureau for Statistics database (www.cbs.nl) to calculate the survival of age- and sex-matched general population groups. Since the mortality rates of the Dutch population varied during the study period, we performed the matching of the cohorts for each year of the study period. We consider this the normal Dutch survival. We compared the late survival of the patient groups with this as the ‘normal Dutch survival’. However, caution in interpreting these results is needed, because the Central Bureau for Statistics database contains the data of the total Dutch population, including patients described in this study as well as the patients who were treated in other Dutch cardiac surgery centres. However, the number of patients in the Netherlands who undergo a CABG in 1 year is relatively small compared with the total number of persons in the general population.

We did not only find that survival rates after CABG in non-diabetic patients were better than in diabetic patients, but also better than the survival rates of the normal Dutch population. An explanation for this might be that a certain percentage of the general population has diabetes and coronary disease. One also may assume that our patients were adequately treated for their coronary problem and postoperatively treated with aspirin and anticholesterolaemic drugs and were treated for underlying diseases such as hypertension. This treatment, as well as the protection obtained by the revascularisation, may contribute to the improved survival and explain why non-diabetic patients have a better survival rate than the general Dutch population. Furthermore, before undergoing CABG, patients are screened for severe underlying disease. If a severe underlying disease is present, an alternative treatment to a CABG must be considered, thus biasing the CABG group.

For both type 1 and type 2 diabetes patients, long-term survival was worse than the normal Dutch survival. Although type 1 diabetes patients were younger, they had survival rates similar to those with type 2 diabetes. The difference in survival between type 1 patients and the normal Dutch population is larger than the difference in survival between type 2 patients and the normal Dutch population indicating a difference between type 1 and type 2 regarding the impact on late survival. This is confirmed by the higher hazard ratio (1.94 for type 1 vs 1.34 for type 2) found by multivariate Cox regression analyses. To our knowledge, this is the first study to compare survival rates of diabetic and non-diabetic CABG patients with age- and sex-matched cohorts from the general population. This information may be valuable for informing patients about their prognosis after a CABG. Instead of informing the patient whether he or she will do better or worse than other patients with or without diabetes undergoing CABG we can inform them how their survival will be compared to the normal survival of people of the same age and sex. For non-diabetic patients, this information strengthens the choice for the standard CABG procedure versus other less invasive options such as PCI. It might be reassuring to know that after their CABG operation, at least for the next 10 years, their survival will be not worse or even better than the normal Dutch survival, unless they have other risk factors for late mortality. For diabetic patients, however, the prognosis, in terms of long-term survival, is worse.

Limitations

This is a retrospective observational study. Therefore, we must be cautious in interpreting our results. The study endpoint is any-cause mortality. We were not able to report causes of death or other morbidities that might be equally interesting. There are no data about perioperative glucose measurement which might affect early outcome.

Conclusions

Only non-insulin-dependent diabetes proved to be a risk factor for early mortality after CABG surgery. Both insulin-dependent and non-insulin-dependent diabetes were predictors of late mortality. Compared with the normal Dutch population, 10-year survival rates were better for non-diabetic patients, while they were worse for non-insulin-dependent diabetic patients and even the worst for insulin-dependent diabetic patients.

References

[1]
Ellis
S.G.
Vandormael
M.G.
Cowley
M.J.G.
Deligonul
U.
Topol
E.J.
Bulle
T.M.
the Multivessel Angioplasty Prognosis Study Group
Coronary morphologic and clinical determinants of procedural outcome with angioplasty for multivessel coronary disease. Implications for patient selection. Multivessel Angioplasty Prognosis Study Group
Circulation
 , 
1990
, vol. 
82
 (pg. 
1193
-
1202
)
[2]
Levine
G.N.
Leya
F.
Keeler
G.P.
Berdan
L.G.
Jacobs
A.K.
the CAVEAT Investigators
Impact of diabetes mellitus on percutaneous revascularization (CAVEAT-I). CAVEAT-I Investigators. Coronary Angioplasty Versus Excisional Atherectomy Trial
Am J Cardiol
 , 
1997
, vol. 
79
 (pg. 
748
-
755
)
[3]
Mueller
H.S.
Lawrence
S.C.
Braunwald
E.
Forman
S.
Feit
F.
Ross
A.
Schweiger
M.
Cabin
H.
Davison
R.
Miller
D.
Solomon
R.
Knatterud
G.L.
Predictors of early morbidity and mortality after thrombolytic therapy of acute myocardial infarction. Analyses of patient subgroups in the Thrombolysis in Myocardial Infarction (TIMI) trial, phase II
Circulation
 , 
1992
, vol. 
85
 (pg. 
1254
-
1264
)
[4]
Alderman
E.L.
Corley
S.D.
Fisher
L.D.
Chaitman
B.R.
Faxon
D.P.
Foster
E.D.
Killip
T.
Sosa
J.A.
Bourassa
M.G.
Five-year angiographic follow-up of factors associated with progression of coronary artery disease in the Coronary Artery Surgery Study (CASS). CASS Participating Investigators and Staff
J Am Coll Cardiol
 , 
1993
, vol. 
22
 (pg. 
1141
-
1154
)
[5]
Gadaleta
D.
Risucci
D.A.
Nelson
R.L.
Tortolani
A.J.
Hall
M.
Parnell
V.
Chiodo
C.
Green
S.
Effects of morbid obesity and diabetes mellitus on risk of coronary artery bypass grafting
Am J Cardiol
 , 
1992
, vol. 
70
 (pg. 
1613
-
1614
)
[6]
Wells
S.
Postoperative recovery: how diabetes complicates care
J Cardiovasc Nurs
 , 
1993
, vol. 
7
 (pg. 
47
-
58
)
[7]
Yamazaki
K.
Kato
H.
Tsujimoto
S.
Kitamura
R.
Diabetes mellitus, internal thoracic artery grafting, and risk of an elevated hemidiaphragm after coronary artery bypass surgery
J Cardiothorac Vasc Anesth
 , 
1994
, vol. 
8
 (pg. 
437
-
440
)
[8]
Slaughter
M.S.
Olson
M.M.
Lee
J.T.
Jr.
Ward
H.B.
A fifteen-year wound surveillance study after coronary artery bypass
Ann Thorac Surg
 , 
1993
, vol. 
56
 (pg. 
1063
-
1068
)
[9]
Stein
B.
Weintraub
W.S.
Gebhart
S.P.
Cohen-Bernstein
C.L.
Grosswald
R.
Liberman
H.A.
Douglas
J.S.
Jr.
Morris
D.C.
King
S.B.
3rd.
Influence of diabetes mellitus on early and late outcome after percutaneous transluminal coronary angioplasty
Circulation
 , 
1995
, vol. 
91
 
4
(pg. 
979
-
989
)
[10]
Smith
L.R.
Harrell
F.E.
Jr.
Rankin
J.S.
Califf
R.M.
Pryor
D.B.
Muhlbaier
L.H.
Lee
K.L.
Mark
D.B.
Jones
R.H.
Oldham
H.N.
Glower
D.D.
Reves
J.G.
Sabiston
D.C.
Jr.
Determinants of early versus late cardiac death in patients undergoing coronary artery bypass graft surgery
Circulation
 , 
1991
, vol. 
84
 
Suppl. 5
(pg. 
III245
-
III253
)
[11]
Calafiore
A.M.
Di Mauro
M.
Di Giammarco
G.
Contini
M.
Vitolla
G.
Iacò
A.L.
Canosa
C.
D’Alessandro
S.
Effect of diabetes on early and late survival after isolated first coronary bypass surgery in multivessel disease
J Thorac Cardiovasc Surg
 , 
2003
, vol. 
125
 
1
(pg. 
144
-
154
)
[12]
Leavitt
B.J.
Sheppard
L.
Maloney
C.
Clough
R.A.
Braxton
J.H.
Charlesworth
D.C.
Weintraub
R.M.
Hernandez
F.
Olmstead
E.M.
Nugent
W.C.
O’Connor
G.T.
Ross
C.S.
Effect of diabetes and associated conditions on long-term survival after coronary artery bypass graft surgery
Circulation
 , 
2004
, vol. 
110
 
11, Suppl. 1
(pg. 
II41
-
II44
)
[13]
Whang
W.
Bigger
J.T.
Jr.
Diabetes and outcomes of coronary artery bypass graft surgery in patients with severe left ventricular dysfunction: results from The CABG Patch Trial database. The CABG Patch Trial Investigators and Coordinators
J Am Coll Cardiol
 , 
2000
, vol. 
36
 
October (4)
(pg. 
1166
-
1172
[Erratum in: J Am Coll Cardiol 2001;37:2012]
[14]
Szabó
Z.
Håkanson
E.
Svedjeholm
R.
Early postoperative outcome and medium-term survival in 540 diabetic and 2239 non-diabetic patients undergoing coronary artery bypass grafting
Ann Thorac Surg
 , 
2002
, vol. 
74
 (pg. 
712
-
719
)
[15]
Barsness
G.W.
Peterson
E.D.
Ohman
E.M.
Nelson
C.L.
DeLong
E.R.
Reves
G.J.
Smith
P.K.
Anderson
R.D.
JonesvF R.H.
Mark
D.B.
Califf
R.M.
Relationship between diabetes mellitus and long-term survival after coronary bypass and angioplasty
Circulation
 , 
1997
, vol. 
96
 (pg. 
2551
-
2556
)
[16]
Clough
R.A.
Leavitt
B.J.
Morton
J.R.
Plume
S.K.
Hernandez
F.
Nugent
W.
Lahey
S.J.
Ross
C.S.
O’Connor
G.T.
The effect of comorbid illness on mortality outcomes in cardiac surgery
Arch Surg
 , 
2002
, vol. 
137
 (pg. 
428
-
432
discussion 432–3
[17]
Niles
N.W.
McGrath
P.D.
Malenka
D.
Quinton
H.
Wennberg
D.
Shubrooks
S.J.
Tryzelaar
J.F.
Clough
R.
Hearne
M.J.
Hernandez
F.
Jr.
Watkins
M.W.
O’Connor
G.T.
Northern New England Cardiovascular Disease Study Group
Survival of patients with diabetes and multivessel coronary artery disease after surgical or percutaneous coronary revascularization: results of a large regional prospective study. Northern New England Cardiovascular Disease Study Group
J Am Coll Cardiol
 , 
2001
, vol. 
37
 (pg. 
1008
-
1015
)
[18]
Detre
K.M.
Guo
P.
Holubkov
R.
Califf
R.M.
Sopko
G.
Bach
R.
Brooks
M.M.
Bourassa
M.G.
Shemin
R.J.
Rosen
A.D.
Krone
R.J.
Frye
R.L.
Feit
F.
Coronary revascularization in diabetic patients: a comparison of the randomized and observational components of the Bypass Angioplasty Revascularization Investigation (BARI)
Circulation
 , 
1999
, vol. 
99
 (pg. 
633
-
640
)
[19]
 
Comparison of coronary bypass surgery with angioplasty in patients with multivessel disease. The Bypass Angioplasty Revascularization Investigation (BARI) Investigators. N Engl J Med 1996; 335:217–25 [Erratum in: N Engl J Med 1997;336:147].
[20]
Herlitz
J.
Wognsen
G.B.
Karlson
B.W.
Sjöland
H.
Karlsson
T.
Caidahl
K.
Hartford
M.
Haglid
M.
Mortality, mode of death and risk indicators for death during 5 years after coronary artery bypass grafting among patients with and without a history of diabetes mellitus
Coron Artery Dis
 , 
2000
, vol. 
11
 (pg. 
339
-
346
)
[21]
Thourani
V.H.
Weintraub
W.S.
Stein
B.
Gebhart
S.S.
Craver
J.M.
Jones
E.L.
Guyton
R.A.
Influence of diabetes mellitus on early and late outcome after coronary artery bypass grafting
Ann Thorac Surg
 , 
1999
, vol. 
67
 (pg. 
1045
-
1052
)
[22]
Carson
J.L.
Scholz
P.M.
Chen
A.Y.
Peterson
E.D.
Gold
J.
Schneider
S.H.
Diabetes mellitus increases short-term mortality and morbidity in patients undergoing coronary artery bypass graft surgery
J Am Coll Cardiol
 , 
2002
, vol. 
40
 (pg. 
418
-
423
)
[23]
Rajakaruna
C.
Rogers
C.A.
Suranimala
C.
Angelini
G.D.
Ascione
R.
The effect of diabetes mellitus on patients undergoing coronary surgery: a risk-adjusted analysis
J Thorac Cardiovasc Surg
 , 
2006
, vol. 
132
 (pg. 
802
-
810
)
[24]
Filsoufi
F.
Rahmanian
P.B.
Castillo
J.G.
Mechanick
J.I.
Sharma
S.K.
Adams
D.H.
Diabetes is not a risk factor for hospital mortality following contemporary coronary artery bypass grafting
Interact Cardiovasc Thorac Surg
 , 
2007
, vol. 
6
 (pg. 
753
-
758
)