University of Groningen Invasive treatment of …Invasive treatment of coronary artery disease: lessons from clinical trials and clinical practice Proefschrift ter verkrijging van

University of Groningen

Invasive treatment of coronary artery diseaseBreeman, Arend

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2006

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Breeman, A. (2006). Invasive treatment of coronary artery disease: lessons from clinical trials and clinicalpractice. s.n.

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 26-06-2020

https://www.rug.nl/research/portal/en/publications/invasive-treatment-of-coronary-artery-disease(d8d22289-9b84-4f7c-93d2-f7f2103d6904).html



Invasive treatment of

coronary artery disease:

lessons from clinical trials

and clinical practice

Arno Breeman

Financial support by the Netherlands Heart Foundation for the publication

of this thesis is gratefully acknowledged.

Financial support by the Zwols Wetenschapsfonds Isala Klinieken - ZWIK

for the publication of this thesis is gratefully acknowledged.

Cover design: Mark Retera / Comic House Oosterbeek

Lay-out: Thea Schenk

ISBN 90-90205-58-6

©2006, A. Breeman. All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of the holder of the copyright.

RIJKSUNIVERSITEIT GRONINGEN

Invasive treatment ofcoronary artery disease:

lessons from clinical trialsand clinical practice

Proefschrift

ter verkrijging van het doctoraat in de Medische Wetenschappen

aan de Rijksuniversiteit Groningen op gezag van de

Rector Magnificus, dr. F. Zwarts, in het openbaar te verdedigen op

maandag 22 mei 2006 om 13:15 uur

door

Arend Breeman geboren op 18 oktober 1960

te Zwijndrecht

Promotor : Prof. dr. F. Zijlstra

Copromotores : Dr. J.P. Ottervanger Dr. E. Boersma

Beoordelingscommissie : Prof. dr. D.J. van Veldhuisen Prof. dr. P. W. Boonstra Prof. dr. M.L. Simoons

Voor Loes, Evert, Karel en Floris

Contents

CHAPTER 1. Introduction and Scope of the thesis ……………………………….. 1 CHAPTER 2. Coronary revascularisation for Non-ST Elevation Acute Coronary Syndrome State of the art ……………………… 11 CHAPTER 3. Completeness of revascularisation by percutaneous coronary intervention ..................................................................... 31 CHAPTER 4. Characteristics, treatment and outcome of patients with non-ST-Elevation acute coronary syndromes and multivessel coronary artery disease: observations from PURSUIT ………………………………………………….………….. 49 CHAPTER 5. Long-term follow-up after invasive approach of coronary artery disease in daily practice ……………………………………… 67 CHAPTER 6. Long-term survival after coronary revascularisation in patients with diabetes ……………………………………………...... 83 CHAPTER 7. Treatment decisions in patients with stable coronary artery disease in a broad range of European practices. Insights from the Euro Heart Survey on Coronary Revascularisation ...………………………………………………….. 99 CHAPTER 8. Diabetes does not influence treatment decisions regarding revascularisation in patients with stable coronary artery disease ……………………………………………. 125

Summary and Conclusions ………………………………………………………….. 149

Samenvatting en Conclusies ...………………………………………………..…….. 161

Curriculum Vitae …………………………...……………………………………….... 173

List of Publications ...……….…………...……………………………………………..175

Dankwoord ……………………………………………………………………………...179

Abbreviations

ACS Acute Coronary Syndrome

ARTS Arterial Revascularization Therapies Study

BARI Bypass Angioplasty Revasularization Investigation

CABG Coronary Artery Bypass Grafting

CABRI Coronary Angioplasty versus Bypass Revascularization

Investigation

CAD Coronary Artery Disease

DES Drug Eluting Stent

DUCAT DUtch inventory of invasive Coronary Atherosclerosis

Treatments

EHS-CR Euro Heart Survey on Coronary Revascularisation

MVD MultiVessel Disease

NSTE-ACS Non-ST Elevation Acute Coronary Syndrome

PCI Percutaneous Coronary Intervention

PURSUIT Platelet Glycoprotein IIb/IIIa in Unstable Angina:

Receptor Suppression Using Integrilin Therapy

RCT Randomised Controlled Trial

STEMI ST-Elevation Myocardial Infarction

CHAPTER 1

Introductionand

Scope of the thesis

Chapter 1 ___________________________________________________________________

3

Introduction

Although there has been a decrease in age-corrected mortality due to

coronary artery disease (CAD) since several decades, coronary heart death

is still the major cause of death in Western societies (figure1) [1]. Despite

this fall in mortality, due to treatment of acute events and primary and

secondary prevention [2], the prevalence of CAD in the Western population

is still very high [3]. Clinical manifestations of CAD include stable angina

and the acute coronary syndromes (ACS), the latter can be classified after

discharge and are divided in Non-ST Elevation Acute Coronary Syndrome

(NSTE-ACS) and ST Elevation Myocardial Infarction (STEMI). Treatment

of CAD consists of lifestyle modification (e.g. diet, smoking cessation,

physical activity), pharmacological treatment (e.g. beta-blockers, nitrates,

statins, antiplatelet agents) and coronary revascularisation by either

percutaneous coronary intervention (PCI) or coronary artery bypass

surgery (CABG).

Cardiovascular (36%)

Cancer (27%)

Respiratory (10%)

Other (27%)

Figure 1: Cause of death in The Netherlands.

Introduction and scope of the thesis ___________________________________________________________________

4

Coronary revascularisation

Since the introduction of CABG in the 60s [4], surgical techniques have

progressed, and care including pharmacotherapy during and after CABG

has improved dramatically. The use of arterial grafts has resulted in better

long-term patency of the grafts [5]. More recently, minimal invasive (‘off

pump’) surgery has been introduced [6]. Also, percutaneous revascular-

isation has evolved after its introduction by Gruentzig in 1977 [7],

particularly by the use of (drug-eluting) stents and medication during and

after the PCI. Nowadays, both CABG and PCI are frequently used and

widely accepted as effective modalities for the treatment of several clinical

manifestations of CAD, with especially an increase in the number of PCIs in

the last 20 years (figure 2).

0

500

1000

1500

2000

PCI CABG Total CR

19912002

Procedures / million population

Figure 2: Coronary revascularisation in Europe [8].

Chapter 1 ___________________________________________________________________

5

Treatment decisions

The aims of treatment of patients with CAD are to improve prognosis by

preventing myocardial infarction, heart failure, arrhythmias or death and to

minimise or abolish (anginal) symptoms. Lifestyle modification is very

important [9,10] and should be advised to all patients. Furthermore,

intensive pharmacological treatment should be optimised in every patient

with CAD. However, whether revascularisation by either CABG or PCI

should be applied in the individual patient may be difficult to decide and

depends of several patient characteristics and the extent of coronary

abnormalities. Furthermore, in individual patients, treatment decisions

should preferably be based on results of randomised trials and

recommendations of guidelines.

Limitations of randomised trials

Randomised controlled clinical trials (RCTs) are the cornerstone of clinical

evidence and recommendations of guidelines. However, it has been

recognised that many of these trials have low external validity and that

patients enrolled in these trials may not reflect those seen in daily clinical

practice [11]. These trials have often only a limited follow-up period.

Another limitation is that effects of treatment may be less clear in several

specific subgroups, because they were not included in the trials (by

exclusion criteria) or because only few patients in that subgroup were

included. Results of registries or observational studies may be more

externally valid than RCTs because patient inclusion is more complete. So,


6

treatment of the individual patient in routine daily practice should be based

on results of both RCTs and observational studies [12].

Sub-groups

Although randomised trials and observational studies may provide

information on the treatment of the general patient, several patient

characteristics may influence treatment decisions. Regarding coronary

revascularisation, apart from comorbidities such as cerebrovascular

disease, renal dysfunction and lung disease, particularly diabetes and the

extent of coronary abnormalities are important. Diabetes has a high

prevalence in patients with CAD. Furthermore, diabetic patients may have a

different response to invasive treatment, and the combination of diabetes

and coronary artery disease has often been associated with a poor

prognosis. There is still debate whether patients with diabetes may profit

more from CABG than PCI [13,14]. Particularly, because the introduction of

newer interventional techniques like stenting has resulted in more

favourable outcome in patients with diabetes [15].

For clinical decision-making, the extent of coronary artery disease is of

course also of importance in choosing medical treatment, PCI or CABG. In

the majority of patients with CAD, more than one coronary vessel is

involved. Randomized trials comparing CABG with PCI for patients with

multivessel disease (MVD) report similar long-term survival for CABG and

PCI [15]. Observational studies, however, suggest survival benefit for

CABG in these patients [16,17], although these analyses included patients

before the era of the drug-eluting stents and the results can be influenced

Chapter 1 ___________________________________________________________________

7

by selection bias. Moreover, these studies give no insight whether

completeness of revascularisation by PCI may improve prognosis.

Scope of the thesis

The main objective of the studies presented in this thesis is to give more

insight in several aspects of coronary revascularisation, including decision-

making in daily practice and (long-term) prognostic implications. Particular

attention is paid to patients with diabetes and multivessel disease.

Chapter 2 reviews current evidence of the prognostic implication of

revascularisation in patients with non-ST elevation acute coronary

syndrome. The results of both randomised trials and observational studies

are summarised, mainly focussing on potential mortality reduction. Also, the

various limitations of both randomised controlled trials and observational

studies are discussed.

The objective in Chapter 3 is to evaluate the potential association between

completeness of revascularisation by percutaneous coronary intervention

and the one-year occurrence of adverse events in patients with multivessel

disease included in the Coronary Angioplasty versus Bypass

Revascularisation Investigation (CABRI) trial. In Chapter 4, characteristics,

treatment and outcome of patients with non-ST elevation acute coronary

syndromes and multivessel coronary artery disease included in the Platelet

Glycoprotein IIb/IIIa in Unstable Angina: Receptor Suppression Using

Integrelin Therapy (PURSUIT) are described. In Chapter 5, the long-term

follow-up after invasive approach of coronary artery disease in daily


8

practice of patients included in the DUtch inventory of invasive Coronary

Atherosclerosis Treatments (DUCAT) study is analysed. It concerns a two-

center follow-up study of all patients consecutively presented to the heart

centers of Zwolle and Groningen during a 3-month period in 1992. In this

study, patients with diabetes have been described in greater detail in

Chapter 6. The Euro Heart Survey (EHS) program was developed to

provide systematic information on the adherence to European and

international guidelines in the member states of the European Society of

Cardiology. In Chapter 7, treatment decisions regarding revascularisation

in patients with stable coronary artery disease in patients enrolled in the

Euro Heart Survey on Coronary Revascularisation (EHS-CR) are analysed.

Chapter 8 presents the analysis of contemporary invasive management

and prognosis of patients with diabetes and stable coronary artery disease

enrolled in the Euro Heart Survey on Coronary Revascularisation (EHS-

CR). It concerns a systematic analysis of the relation between patient

characteristics (including diabetes), invasive treatment choices and

prognosis.

Chapter 1 ___________________________________________________________________

9

References

1. www.heartstats.org.2. Unal B, Critchley JA, Capewell S. Modelling the decline in coronary heart disease deaths in England and Wales, 1981-2000: comparing contributions from primary prevention and secondary prevention. BMJ 2005;331:614-9. 3. McGill HC, McMahan A, Zieske AW, et al. Association of Coronary Heart Disease Risk Factors with microscopic qualities of coronary atherosclerosis in youth. Circulation 2000;102:374-9. 4. Mueller RL, Rosengart TK, Isom OW. The history of surgery for ischemic heart disease. Ann Thorac Surg 1997;63:869-78. 5. Bergsma TM, Grandjean JG, Voors AA, Boonstra PW, den Heyer P, Ebels T. Low recurrence of angina pectoris after coronary artery bypass graft surgery with bilateral internal thoracic and right gastroepiploic arteries. Circulation 1998;97:2402-5. 6. van der Heijden GJ, Nathoe HM, Jansen EW, Grobbee DE. Meta-analysis on the effect of off-pump coronary bypass surgery. Eur J Cardiothorac Surg 2004;26:81-4. 7. Gruentzig A. Transluminal dilatation of coronary artery stenosis. Lancet 1978;1:263. 8. Ghosh P, Schistek R, Unger F. Coronary revascularisation in DACH: 1991- 2002. Thorac Cardiovasc Surg 2004;52:356-64. 9. Yusuf S, Hawken S, Ounpuu S, et al. Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): case-control study. INTERHEART study investigators. Lancet 2004;364:937- 52. 10. Knoops KT, Groot LC de , Kromhout D, et al. Mediterranean diet, lifestyle factors, and 10-year mortality in elderly European men and women: the HALE project. JAMA 2004;292:1433-9. 11. Rothwell PM. External validity of randomised controlled trials: “To whom do the results of this trial apply?” Lancet 2005;365:82–93. 12. McKee M, Britton A, Black N, McPherson K, Sanderson C, Bain C. Methods in health service research. Interpreting the evidence: choosing between randomised and non-randomised studies. BMJ 1999; 319: 312–15. 13. Detre KM, Guo P, Holubkov R, et al. Coronary revascularisation in diabetic patients. A comparison of the randomized and observational components of the Bypass Angioplasty Revascularisation Investigation (BARI). Circulation 1999;99:633-640. 14. Feit F, Brooks MM, Sopko G, et al. Long-term clinical outcome in bypass angioplasty revascularisation investigation registry. Comparison with the randomised trial. Circulation 2000;101:2795-2802.


10

15. Serruys PW, Ong ATL, Van Herwerden LA, et al. Five year outcome after coronary stenting versus bypass surgery for the treatment of multivessel disease. J Am Coll Cardiol 2005;46:575-81. 16. Malenka DJ, Leavitt BJ, Hearne MJ, et al. Comparing long-term survival of patients with multivessel coronary disease after CABG or PCI: analysis of BARI-like patients in northern New England. Circulation 2005;112(Suppl):- I371-6. 17. Hannan EL, Racz MJ, Walford G, et al. Long-term outcomes of coronary-artery bypass grafting versus stent implantation. N Engl J Med 2005;352:2174-83.

CHAPTER 2

Coronary revascularisation for Non-ST

Elevation Acute Coronary Syndrome

State of the art

A. Breeman, J.P. Ottervanger

E. Boersma, G. De Luca

J Cardiovascul Med (in press)

Chapter 2 ___________________________________________________________________

13

Abstract

Immediate as well as early revascularisation may be beneficial in patients

with acute coronary syndromes (ACS) without ST elevation, but has

traditionally employed as an ‘ischemia-guided’ strategy. A number of

randomised trials (including more than 10,000 patients) and observational

studies have compared routine invasive versus selective invasive treatment

in patients with an acute coronary syndrome without ST elevation. Most

randomised trials are limited by a high cross-over rate, whereas

observational studies are limited by selection bias. Data from registries

demonstrated benefits with an invasive approach. Results from randomised

trials are less clear regarding mortality reduction, although long-term

survival after hospital discharge may be better after the invasive approach.

In the randomised trials there was a decreased risk of death or MI after the

invasive approach (OR 0.88, 95% CI 0.76-1.0). Despite the optimal timing

of angiography and subsequent revascularisation, if appropriate, may be

uncertain, angiography and revascularisation should be considered in every

patient admitted with an ACS without ST elevation, particularly in patients

with high-risk characteristics, such as ST segment depression >0.1 mV,

accelerated angina in the prior 2 months, nitrate use in the prior week

before admission and elevated troponin.

Introduction

Prognosis of patients with stable angina is in general very good, with an

incidence of death or non-fatal myocardial infarction not exceeding 2% per

Revascularisation for NSTE-ACS ___________________________________________________________________

14

year [1]. In contrast, patients with an acute coronary syndrome (ACS)

without ST elevation (NSTE-ACS) have a much worse prognosis, as 10-

15% experience death or non-fatal myocardial infarction within one year

after admission [2]. According to the European and American treatment

guidelines, and based on a large number of randomised trials, medical

therapy of NSTE-ACS patients should preferably include heparin,

antiplatelet agents and anti-ischemic agents [3]. Coronary revascularisation

by either percutaneous coronary intervention (PCI) or coronary bypass

surgery (CABG) is recommended in patients with recurrent ischemia, as

well as in those who are otherwise at increased risk. In clinical practice,

however, an important limiting factor to offer coronary revascularisation is

its availability [4]. Patients presenting to hospitals without facilities for

coronary revascularisation are much less likely to receive such treatment

than those who present to hospitals with more extensive equipment. A

number of registries and randomised trials describe potential benefits of

(early) revascularisation in NSTE-ACS patients. We aimed to review these

data.

Randomised trials

Between 1994 and 2002 eight clinical trials were undertaken, which

randomised approximately 9000 NSTE-ACS patients to an early invasive or

conservative treatment strategy [5-11]. Recently, results of an additional

trial (ICTUS) were published [12]. In most trials, patients randomised to

invasive treatment underwent cardiac catheterisation during the first days

after hospital admission (in the majority of patients cardiac catheterisation

Chapter 2 ___________________________________________________________________

15

was performed within 72 h), and, if necessary, revascularisation as soon as

possible thereafter. Treatment decisions in patients randomised to the

conservative treatment strategy were left to the discretion of the physician,

and were mainly based on symptoms. All trials together, early invasive

treatment was not associated with a mortality reduction after one-year

follow-up (table 1; figure 1 left panel), although in a recently published

meta-analysis it was shown that after hospital discharge the invasive

approach is associated with mortality reduction after long-term follow-up

[13]. Early invasive treatment was associated with a borderline statistically

significant 12% relative reduction – and a 1.1% absolute reduction – in the

composite endpoint of death or non-fatal myocardial infarction during long-

term follow-up compared with conservative treatment (8.7% versus 9.8%

events; odds ratio 0.88 and 95% CI 0.77-1.0). However, there was

evidence of a differential treatment effect between the separate trials.

Figure 1: Mortality (left panel) and the composite endpoint of death or non-fatal MI (right panel) in the randomised trials comparing invasive with conservative approach for non-ST elevation ACS.


16

Tabl

e 1:

Inci

denc

e of

dea

th a

nd n

on-fa

tal m

yoca

rdia

l inf

arct

ion

durin

g on

e-ye

ar fo

llow

-up

in ra

ndom

ised

tria

ls o

f in

vasi

ve v

ersu

s co

nser

vativ

e tre

atm

ent i

n pa

tient

s pr

esen

ting

with

NS

TE-A

CS

.

Chapter 2 ___________________________________________________________________

17

Particularly, divergent results were observed in the VANQWISH trial, as

patients randomised to early invasive treatment had significant higher 1-

year event rates (22.1% versus 14.2% events; odds ratio 1.7 and 95% CI

1.2-2.4) [6]. The VANQWISH results, however, have been challenged,

since a relative large proportion of patients allocated to the conservative

strategy actually underwent coronary revascularisation within 30 days after

randomisation (33% compared with 44% in the early invasive arm).

Furthermore, in the earlier trials, among which VANQWISH, neither

intracoronary stents nor GP IIb/IIIa inhibitors were available for patients

undergoing PCI. Finally, in VANQWISH an exceptionally high peri-operative

mortality was observed in the invasive arm (11.6% versus 3.4% events;

odds ratio 3.7 and 95% CI 1.0-13.6). A meta-analysis of the five recent

trials revealed a 28% relative reduction – and a 2.5% absolute reduction –

in the composite endpoint of death or non-fatal myocardial infarction by the

early invasive strategy (7.0% versus 9.5% events; odds ratio 0.72 and 95%

CI 0.60-0.86).

In ICTUS, a total of 1201 NSTE-ACS patients who had a cTnT

concentration >0.03 μg/L at presentation, were randomly assigned to a

treatment strategy that aimed at coronary angiography and subsequent

revascularisation within 24-48 hrs (early invasive approach) or to a

treatment strategy in which coronary angiography (and revascularisation)

was only performed in cases of refractory angina or ischemia on

predischarge exercise testing (selective invasive approach). In both

treatment arms abciximab was given whenever a PCI was performed,

whereas coronary stents were used liberally. Within one year after

randomisation, coronary revascularisation was performed in 73% of


18

patients allocated to early invasive treatment versus 47% in those allocated

to selective invasive treatment, whereas timing of revascularisation was

within 30 days in almost all patients who had revascularisation. The early

invasive approach was associated with an increased incidence of non-fatal

myocardial infarction during one-year follow-up compared to the selective

invasive approach (15% versus 10%; odds ratio 1.5 and 95% CI 1.1 – 2.0),

but most events were related to the revascularisation. Patients allocated to

the invasive approach had a lower incidence of rehospitalisation for ACS

within one year (7.4% and 10.9%; odds ratio 0.68, 95% CI 0.47-0.98).There

was no difference in all-cause mortality after one year between the

separated approaches (2.5% versus 2.5%; odds ratio 0.99 and 95% CI

0.49-2.00). Compared to the other randomised trials, patients allocated to

conservative treatment had a high frequency of cross-over to (early)

revascularisation (47%) and a low one-year mortality (2.5% in ICTUS vs.

4.4% in the other trials).

Apart from potential effects on mortality and recurrent MI, in the

randomised trials it was shown that early revascularisation may particularly

reduce anginal complaints and rehospitalisation [13]. Also in ICTUS, the

incidence of rehospitalisation for anginal complaints was significantly lower

in those allocated to the early invasive strategy.

Randomised controlled trials are the cornerstone of evidence based

medicine. Still, clinical trials comparing invasive versus medical approach in

NSTEMI should be interpreted with caution. Importantly, in these trials,

many patients allocated to conservative treatment crossed-over to invasive

treatment, whereas many patients allocated to invasive treatment did not

undergo such therapy. Furthermore, patients included in these trials consist

Chapter 2 ___________________________________________________________________

19

a selected patient group, with differences to patients seen in daily clinical

practice. Also, these trials cannot be blinded to treatment allocation. These

factors may have caused bias, and particularly cross-over in the

conservative group may have contributed to an underestimation of the

benefits with an invasive approach in NSTE-ACS.

Observational studies

Apart from the randomised trials, there are several observational studies,

including registries, regarding the potential benefit of revascularisation in

patients with ACS without ST elevation [14-19]. In the OASIS registry,

evaluating approximately 8000 patients with suspected NSTE-ACS from 6

countries, a reduction in refractory ischemia and need for hospitalisation

was observed after revascularisation, but not a decrease of mortality [16].

Analyses of data from patients included in GUSTO IV-ACS provided

evidence that coronary revascularisation within 30 days after admission

was associated with a favourable one-year survival, with one-year mortality

2.3% in the invasive group vs 5.6% in the conservative group [17]. This

beneficial effect of revascularisation was also observed after adjusting for

several baseline characteristics and a propensity score: patients with

revascularisation had a relative risk of one-year mortality of 0.53 (95% CI

0.37-0.77) compared to patients without revascularisation. Moreover, in this

analysis, survival benefit after revascularisation, independent of early risk

stratification, was consistently shown in every subgroup studied.


20

The CRUSADE (Can Rapid Risk Stratification of Unstable Angina Patients

Suppress Adverse Outcomes With Early Implementation of the ACC/AHA

Guidelines) Quality Improvement Initiative evaluated care patterns and

outcomes of 17,926 high-risk NSTE ACS patients with positive cardiac

markers and/or ischemic electrocardiographic changes at admission [18].

Patients treated with early invasive management had a lower risk of in-

hospital mortality after adjusting for differences in clinical characteristics

and after comparing propensity-matched pairs (2.5% vs 3.7%, P<0.001,

odds ratio 0.67; 95% CI 0.56-0.80).

In summary, in most observational studies benefits of early

revascularisation appeared larger than in the randomised trials, even after

adjustment for several patient and hospital characteristics. It should be

appreciated, nevertheless, that such multivariable adjustments may not

correct for all patient characteristics, and does not fully compensate the

effect of careful selection of patients who are candidates for

revascularisation, and those who are not. The true benefit of

revascularisation might probably be overestimated by registries and

underestimated by randomised trials.

Revascularisation in practice

Recent European and American guidelines for clinical practice advise an

invasive strategy in patients with ACS and failed medical therapy as well in

high risk patients [3,20]. This is in agreement with observations from FRISC

II and TACTICS-TIMI 18 demonstrating that the absolute reduction of death

or subsequent myocardial infarction is highest in high-risk groups. Failure of

Chapter 2 ___________________________________________________________________

21

medical therapy include ST segment depression >0.1 mV, accelerated

angina in the prior 2 months, nitrate use in the prior week before admission

and elevated troponin [21]. Whether other diagnostic tools, such as

(contrast) echocardiography may be helpful in selecting patients for an

early invasive approach should be assessed [22,23]. Finally, although a

number of studies suggest benefits of revascularisation, data from

registries showed that in daily clinical practice an early invasive

management strategy is commonly denied in many high-risk patients with

NST-ACS [18,24-25].

Timing of revascularisation

The hazard of an unfavourable event in patients with NST-ACS is highest

within the first 48 hours after admission [26]. Nevertheless, the appropriate

timing of coronary angiography and revascularisation remains

controversial. In an early invasive approach, the risk of peri-procedural

complications may be increased [27]. Therefore, a period of ‘cooling off’, in

which patients are medically stabilised, has been recommended. This

allows partial resolution of the initially strong inflammatory response with

liberation of cytokines [28], increased generation of free radicals [29], and

enhanced activity of the coagulation system [30]. On the other hand, early

angiography and subsequent revascularisation may prevent myocardial

infarction or death in a high risk period, particularly with the use of stents

and GP IIb/IIIa inhibitors. In PURSUIT, early PCI (within 24 h) in patients

receiving a GP IIb/IIIa inhibitor had the lowest event rate after 30 days [31].

In TACTICS TIMI 18, there was no mortality difference between patients


22

with revascularisation within 48 hours after randomisation, and those with

revascularisation after 48 hours [32]. In GUSTO IV-ACS, an increased

mortality was found in patients with early CABG [17]. Possibly, when

indicated, PCI should be performed, as soon as possible after admission

for ACS while CABG should be deferred. However, until now, there are only

few randomised data available, comparing an early with a later invasive

approach, showing conflicting results [33,34].

Angioplasty or surgery

In the randomised trials of invasive versus conservative management in

ACS, there are different findings with regard to mortality after CABG. In

FRISC II the in-hospital and 30-day mortality after CABG in the invasive

group was respectively 1.2% and 2.1%. Similar low rates were observed in

TACTICS TIMI 18, and outcome was comparable between patients

undergoing CABG and PCI. In VANQWISH, however, there was a very

high mortality in the CABG group. In recent randomised trials comparing

stented angioplasty and bypass surgery, which included patients with non-

STEMI, the incidence of death or myocardial infarction during follow-up of 2

years were similar between the two groups [35,36]. The less favourable

outcome after CABG in GUSTO IV-ACS was most likely related to the

selection of patients with more extensive disease to undergo surgery. Data

of randomised studies with drug-eluting stents are not yet available.

Whether complete revascularisation may reduce mortality compared to

revascularisation of the culprit lesion is unknown. A small study showed no

Chapter 2 ___________________________________________________________________

23

mortality difference, but a trend to fewer readmissions for unstable angina

and less long-term antianginal medication [37].

GP IIb/IIIa inhibitors

A number of previous studies have demonstrated beneficial effects of GP

IIb/IIIa inhibitors in patients with ACS without ST elevation, in particular in

patients undergoing PCI [38-41]. In the guidelines of management of

patients with non-STEMI, use of GP IIb/IIIa inhibitors are recommended, in

addition to standard treatment (aspirin and unfractionated heparin or low-

molecular-weight heparins) for patients with high-risk features such as

elevated troponin, ST-segment changes, or recurrent ischaemia [3,42]. GP

IIb/IIIa inhibitors are particularly effective in patients with NSTE-ACS when

early PCI is performed [31]. GUSTO IV-ACS did not show beneficial effects

of abciximab [17]. However, in this trial, coronary angiography was

discouraged within 60 h after randomisation unless the patient had

recurrent or continuing ischemia at rest associated with ischemic ST-T

changes that were unresponsive to intensive medical therapy. Indeed, only

128 patients (1.6%) had PCI during study drug infusion.

In summary, in patients with non-ST-segment elevation ACS and early

invasive management, optimal pharmacological therapy should be

supported, including GP IIb/IIIa antagonists [43]. However, in women and

troponin-negative patients data are less clear and enhanced risk

stratification is needed in these subgroups [44]. Furthermore, all previous

large trials on GP IIb/IIIa inhibitors during PCI for non-ST ACS have been

performed without routine use of clopidogrel. A relative small study was


24

recently published, and suggested no benefit of routine GP IIb/IIIa inhibitors

compared to provisional use in patients with non-ST ACS treated with

aspirin, heparin and clopidogrel [45].

Sub-groups of patients

There are several patient characteristics that should be considered in

patient selection for revascularisation. Although increasing age is an

important risk factor for adverse outcome among patients with acute

coronary syndromes, elderly patients are more often managed

conservatively. A sub-analysis of patients of the TACTICS-TIMI 18 trial

showed that, despite an increased risk for major bleeding in patients older

than 75 years of age, a routine early invasive strategy can significantly

improve ischemic outcomes in elderly patients with unstable angina and

non-ST segment elevation MI [46]. These data have been confirmed in

other studies [47,48].

In patients admitted with NSTE-ACS, females are often older and have

more often diabetes mellitus or hypertension [49-51]. It has been suggested

that GP IIb/IIIa inhibitors are not beneficial in females with unstable angina

[52] and that invasive treatment is less effective in females compared to

males [53,54]. This was, however, not confirmed by other studies [55]. Also

data from GUSTO IV-ACS did not show significant differences between

males and females regarding mortality reduction of revascularisation [17].

In most studies, diabetic patients admitted with a non-ST elevation ACS are

older and more likely to be female than non-diabetic patients. They have

Chapter 2 ___________________________________________________________________

25

more previous cardiovascular events, including MI, congestive heart failure,

stroke, and revascularisation procedures [56]. Furthermore, diabetes is an

independent predictor of mortality after a non-ST elevation ACS [56-58].

Coronary revascularisation in patients with diabetes and non-ST elevation

ACS may be, however, as effective as in patients without diabetes [59].

Conclusions

Although the results observed in randomised trials are less clear than

observational studies, we believe that angiography and revascularisation

should be strongly considered during the initial hospitalisation in high-risk

patients with NSTE-ACS and in case of failure of medical therapy in

suppressing ischemia. The optimal timing of angiography and subsequent

revascularisation if appropriate remains uncertain, but possibly PCI should

be performed as soon as possible after admission for ACS, while CABG

should be deferred. We did not review data on benefits of revascularisation

apart from survival or recurrent MI, but patients have probably less com-

plaints during follow-up after revascularisation, as was shown in RITA 3.


26

References

1. Juul-Möller S, Edvardsson N, Jahnmatz B, Rosen A, Sørensen S, Ömblus R, for the Swedish Angina Pectoris Aspirin Trial (SAPAT) Group. Double-blind trial of aspirin in prevention of myocardial infarction in patients with stable chronic angina pectoris. Lancet 1992;340:1421-5. 2. Yeghiazarians Y, Braunstein JB, Askari A, Stone P. Unstable angina pectoris. N Engl J Med 2000;342;101-14. 3. Bertrand ME, Simoons ML, Fox KA, Wallentin LC, Hamm CW, McFadden E. Task Force on the Management of Acute Coronary Syndromes of the European Society of Cardiology. Management of acute coronary syndromes in patients presenting without persistent ST-segment elevation. Eur Heart J 2002;23:1809-40. 4. Hasdai ED, Behar S, Boyko V, Danchin N, Bassand JP, Battler A. Cardiac biomarkers and acute myocardial syndromes – the Euro Heart Survey of Acute Coronary Syndromes Experience. Eur Heart J 2003;24:1189-94. 5. TIMI-III investigators. Effects of tissue plasminogen activator and a comparison of early invasive and conservative strategies in unstable angina and non-Q- wave myocardial infarction: results of the TIMI-IIIB trial. Circulation 1994;89:1545-56. 6. Boden WE, O’Rourke RA, Crawford MH, et al. Outcomes in patients with acute non-Q wave myocardial infarction randomly assigned to an invasive as compared to a conservative strategy. N Engl J Med 1998;338:1785-92. 7. Wallentin L, Lagerqvist B, Husted S, et al. Outcome at 1 year after an invasive compared with a non-invasive strategy in unstable coronary-artery disease: the FRISC II invasive randomised trial. Lancet 2000;356:9-16. 8. Cannon C. Tactics-TIMI 18 (Treat angina with Aggrastat and determine cost of therapy with an invasive or conservative strategy). N Engl J Med 2001;344:1879-87. 9. Michalis LK, Stroumbis CS, Pappas K, et al. Treatment of refractory unstable angina in geographically isolated areas without cardiac surgery. Invasive versus conservative strategy (TRUCS Study). Eur Heart J 2000;21:1954-9. 10. Spacek R, Widimsky P, Straka Z, et al. Value of first day angiography/angioplasty in evolving Non-ST segment elevation myocardial infarction: an open multicenter randomized trial. The VINO Study. Eur Heart J 2002;23:230-8. 11. Fox KA, Poole-Wilson PA, Henderson RA, et al. Interventional versus conservative treatment for patients with unstable angina or non-ST-elevation myocardial infarction: the British Heart Foundation RITA 3 randomised trial. Randomized Intervention Trial of unstable Angina. Lancet 2002;360:743-51. 12. Winter de RJ, Windhausen F, Cornel JH, et al. for the ICTUS investigators. Early invasive versus selectively invasive management for acute coronary syndromes. N Engl J Med 2005;353:1095-104.

Chapter 2 ___________________________________________________________________

27

13. Mehta SR, Cannon CP, Fox KAA, et al. Routine vs selective invasive strategies in patients with acute coronary syndromes. A collaborative meta- analysis of randomized trials. JAMA 2005;293:2908-17. 14. Shavelle DM, Parsons L, Sada MJ, French WJ, Every NR. Is there a benefit to early angiography in patients with ST-segment depression myocardial infarction? An observational study. Am Heart J 2002;143:488-96. 15. Scull GS, Martin JS, Weaver WD, Every NR. Early angiography versus conservative treatment in patients with non-ST elevation acute myocardial infarction: MITI investigators. Myocardial Infarction Triage and Intervention. J Am Coll Cardiol 2000;35:903-6. 16. Yusuf S, Flather M, Pogue J, et al. Variations between countries in invasive cardiac procedures and outcome in patients with suspected unstable angina or myocardial infarction without initial ST elevation. Lancet 1998;352:507-14. 17. Ottervanger JP, Armstrong P, Barnathan ES, et al. Association of revascularisation with low mortality in Non-ST Elevation Acute Coronary Syndrome, a report from GUSTO IV-ACS. Eur Heart J 2004;25:1494-501. 18. Bhatt DL, Roe MT, Peterson ED, et al. Utilization of early invasive management strategies for high-risk patients with non-ST-segment elevation acute coronary syndromes: results from the CRUSADE Quality Improvement Initiative. JAMA 2004;292:2096-104. 19. Scull GS, Martin JS, Weaver WD, Every NR. Early angiography versus conservative treatment in patients with non-ST elevation acute myocardial infarction: MITI Investigators. Myocardial Infarction Triage and Intervention. J Am Coll Cardiol 2000;35:895-902. 20. Braunwald E, Raymond R, Antman EM, et al. ACC/AHA 2002 Guidelines update for the management of patients with unstable angina and non-ST segment elevation myocardial infarction-summary article. J Am Coll Cardiol 2002;40:1366-74. 21. Solomon DH, Ganz DA, Avorn J, et al. Which patients with unstable angina or non-Q-wave myocardial infarction should have immediate cardiac catheterization? A clinical decision rule for predicting who will fail medical therapy. J Clin Epidem 2002;55:121-8. 22. Tong KL, Kaul S, Wang XQ, et al. Myocardial contrast echocardiography versus Thrombolysis In Myocardial Infarction score in patients presenting to the emergency department with chest pain and a nondiagnostic electrocardiogram. J Am Coll Cardiol 2005;46:920-7. 23. Kaul S, Senior R, Firschke C, et al. Incremental value of cardiac imaging in patients presenting to the emergency department with chest pain and without ST-segment elevation: a multicenter study. Am Heart J 2004;148:129-36. 24. Collinson J, Flather MD, Fox KA, et al. Clinical outcomes, risk stratification and practice patterns of unstable angina and myocardial infarction without ST elevation: Prospective Registry of Acute Ischaemic Syndromes in the UK (PRAIS-UK). Eur Heart J 2000;17:1450-7.


28

25. Ellis C, Devlin G, Matsis P, et al., for the New Zealand Acute Coronary Syndromes [NZACS] Audit Group. Acute Coronary Syndrome patients in New Zealand receive less invasive management when admitted to hospitals without invasive facilities. NZ Med J 2004;11:U954. 26. Cohen M, Antman E, Gurfinkel E, Turpie AGG, Bigonzi F, Radley D. Timing of events after hospitalization for non-ST segment elevation acute coronary syndromes. Circulation 1999;100(Suppl I):I-641. 27. De Feyter PJ, Serruys PW: Percutaneous transluminal coronary angioplasty for unstable angina. In: Textbook of Interventional Cardiology, Vol 1, 2nd edn. Edited by Topol EJ. Philadelphia, PA: WB Saunders, 1993;274-91. 28. Liuzzo G, Buffon A, Biasucci LM, et al. Enhanced inflammatory response to coronary angioplasty in patients with severe unstable angina. Circulation 1998;98:2370-6. 29. Buffon A, Santini SA, Ramazzotti, et al. Large, sustained cardiac lipid peroxidation and reduced antioxidant capacity in the coronary circulation after brief episodes of myocardial ischemia. J Am Coll Cardiol 2000;35:633-9. 30. Fuster V, Fayad Z, Badimon J. Acute coronary syndromes: biology. Lancet 1999;SII:5-9. 31. Ronner E, Boersma E, Akkerhuis KM, et al. Patients with acute coronary syndromes without persistent ST elevation undergoing percutaneous coronary intervention benefit most from early intervention with protection by a glycoprotein IIb/IIIa receptor blocker. Eur Heart J 2002;23:239-46. 32. McCullough PA, Gibson CM, Dibattiste PM, et al. Timing of angiography and revascularization in acute coronary syndromes: an analysis of the TACTICS- TIMI-18 trial. J Interv Cardiol 2004;17:81-6. 33. van ’t Hof AWJ, de Vries ST, Dambrink JH, et al. A comparison of two invasive strategies in patients with non-ST elevation acute coronary syndromes: results of the Early or Late Intervention in unStable Angina (ELISA) pilot study. 2b/3a upstream therapy and acute coronary syndromes. Eur Heart J 2003;24:1401-5. 34. Neumann FJ, Kastrati A, Pogatsa-Murray G, et al. Evaluation of prolonged antithrombotic pretreatment (“cooling-of” strategy) in patients with unstable coronary syndromes: a randomized controlled trial. JAMA 2003 ;290 :1593-9. 35. de Feyter PJ, Serruys PW, Unger F, et al. Bypass surgery versus stenting for the treatment of multivessel disease in patients with unstable angina compared with stable angina. Circulation 2002;105:2367-72. 36. Stables R, for the SoS Investigators. Coronary artery bypass surgery versus percutaneous coronary intervention with stent implantation in patients with multivessel coronary artery disease (the Stent or Surgery trial): a randomised controlled trial. Lancet 2002;360:965-70. 37. Palmer ND, Causer JP, Ramsdale DR, Perry RA. Effect of completeness of revascularization on clinical outcome in patients with multivessel disease presenting with unstable angina who undergo percutaneous coronary intervention. J Invasive Cardiol 2004;16:185-8.

Chapter 2 ___________________________________________________________________

29

38. CAPTURE Investigators. Randomised placebo-controlled trial of abciximab before and during coronary intervention in refractory unstable angina: The CAPTURE study. Lancet 1997;349:1429-35. 39. PURSUIT Trial Investigators. Inhibition of platelet glycoprotein IIb/IIIa with eptifibatide in patients with acute coronary syndromes. N Engl J Med 1998;339:436-43. 40. PRISM-PLUS Study Investigators. Inhibition of the platelet glycoprotein IIb/IIIa receptor with tirofiban in unstable angina and non-Q-wave myocardial infarction. Platelet Receptor Inhibition in Ischemic Syndrome Management in Patients Limited by Unstable Signs and Symptoms (PRISM-PLUS) Study Investigators. N Engl J Med 1998;338:1488-97. 41. PARAGON Investigators. International, randomized, controlled trial of lamifiban (a platelet glycoprotein IIb/IIIa inhibitor), heparin, or both in unstable angina. Circulation 1998;97:2386-95 42. Hamm CW, Bertrand M, Braunwald E. Acute coronary syndrome without ST elevation: implementation of new guidelines. Lancet 2001;358:1533-8. 43. Atwater BD, Roe MT, Mahaffey KW. Platelet glycoprotein IIb/IIIa receptor antagonists in non-ST segment elevation acute coronary syndromes: a review and guide to patient selection. Drugs 2005;65:313-24. 44. Bavry AA, Kumbhani DJ, Quiroz R, Ramchandani SR, Kenchaiah S, Antman EM. Invasive therapy along with glycoprotein IIb/IIIa inhibitors and intracoronary stents improves survival in non-ST-segment elevation acute coronary syndromes: a meta-analysis and review of the literature. Am J Cardiol 2004;93:830-5. 45. Leoncini M, Toso A, Maioli M, et al. Effects of tirofiban plus clopidogrel versus clopidogrel plus provisional abciximab on biomarkers of myocardial necrosis in patients with non-ST-elevation acute coronary syndromes treated with early aggressive approach. Results of the CLOpidogrel, upstream Tirofiban, in cath Lab Downstream Abciximab (CLOTILDA) study. Am Heart J 2005;150:401-3. 46. Bach RG, Cannon CP, Weintraub WS, et al. The effect of routine, early invasive management on outcome for elderly patients with non-ST-segment elevation acute coronary syndromes. Ann Intern Med 2004;141:186-95. 47. Leoncini M, Maioli M, Bellandi F, et al. Therapeutic strategies, immediate and mid-term outcomes in non-ST-segment elevation acute coronary syndromes with respect to age: a single-center registry of 488 consecutive patients. Clin Cardiol 2004;27:475-9. 48. Blackman DJ, Ferguson JD, Sprigings DC, Banning AP. Revascularization for acute coronary syndromes in older people. Age Ageing 2003;32:129-35. 49. Chua TP, Saia F, Bhardwaj V, et al. Are there gender differences in patients presenting with unstable angina? Int J Cardiol 2000;72:281-6. 50. Scirica BM, Moliterno DJ, Every NR, et al. Differences between men and women in the management of unstable angina pectoris (The GUARANTEE Registry). Am J Cardiol 1999;84:1145-50.


30

51. Hochman JS, mcCabe CH, Stone PH, et al. Outcome and profile of women and men presenting with acute coronary syndromes: A report from TIMI IIIB. J Am Coll Cardiol 1997;30:141-8. 52. Boersma E, Harrington RA, Moliterno DJ, White H, Thérous P, Van de Werf F, et al. Platelet glycoprotein IIb/IIIa inhibitors in acute coronary syndromes. A meta-analysis of all randomised clinical trials that enrolled over 1000 patients. Lancet 2002;359:189-98. 53. Lagerqvist B, Säfstrom K, Ståhle E, Wallentin L, Swahn E, et al. Is early invasive treatment of unstable coronary artery disease equally effective for both women and men? J Am Coll Cardiol 2001;38:41-8. 54. Gowda MS, Vacek JL, Hallas D. Gender-related risk factors and outcomes for non-Q wave myocardial infarction patients receiving in-hospital PTCA. J Invasive Cardiol 1999;11:1212-6. 55. Glaser R, Herrmann HC, Murphy SA, et al. Benefit of an early invasive management strategy in women with acute coronary syndromes. JAMA. 2002;288:3124-9. 56. Malmberg K, Yusuf S, Gerstein HC, et al. Impact of diabetes on long-term prognosis in patients with unstable angina and non-Q-wave myocardial infarction: results of the OASIS (Organization to Assess Strategies for Ischemic Syndromes) Registry. Circulation. 2000;102:1014-9. 57. Svensson AM, Abrahamsson P, McGuire DK, Dellborg M. Influence of diabetes on long-term outcome among unselected patients with acute coronary events. Scand Cardiovasc J. 2004;38:229-34. 58. Franklin K, Goldberg RJ, Spencer F, et al. for the GRACE Investigators. Implications of diabetes in patients with acute coronary syndromes. The Global Registry of Acute Coronary Events. Arch Intern Med. 2004;164:1457- 63. 59. Norhammar A, Malmberg K, Diderholm E, et al. Diabetes mellitus: the major risk factor in unstable coronary artery disease even after consideration of the extent of coronary artery disease and benefits of revascularization. J Am Coll Cardiol 2004;43:585-91.

CHAPTER 3

Completeness of revascularisation by

percutaneous coronary intervention

A. Breeman, E. Boersma, M.J. van den Brand,

L. van Herwerden, P.W. Serruys

for the CABRI-investigators

Neth Heart J 2001;9:3-9

Chapter 3___________________________________________________________________

33

Abstract

Objective: To evaluate the relationship between the completeness of

revascularisation by percutaneous coronary intervention and the one-year

occurrence of adverse cardiac events in patients with multivessel disease.

Patients. Patients with stable or unstable angina pectoris, or with exercise-

induced ischaemia, were enrolled in the Coronary Angioplastry versus

Bypass Revascularisation Investigation (CABRI).

Methods: In CABRI, patients were randomised to coronary bypass grafting

(CABG; n=513) or angioplasty (PTCA; n=541). Revascularisation in

patients randomised to PTCA was defined as complete if no lesions with a

diameter stenosis <50% remained post-procedure. Patients with complete

revascularisation were distinguished from those with one, two or three or

more remaining lesions, respectively. Differences in baseline

characteristics and in the one-year occurrence of death, myocardial

infarction, (re)CABG and (re)PTCA between these subgroups were

evaluated. Comparisons were made with patients randomised to CABG.

Results: Complete revascularisation was obtained in 148 patients

randomised to PTCA (27%). In 147 (27%) cases one lesion remained,

while there were 122 (23%) and 119 (22%) patients with two and three or

more remaining lesions, respectively. Five (1%) patients could not be

classified. The one-year rates of either death or MI were 9.5%, 5.4%, 8.2%

and 12.6% in the respective PTCA subgroups (p=0.225), and 6.2% in

patients randomised to CABG (comparison with three or more remaining

lesions after PTCA: p=0.017) The percentages of repeat interventions

during one-year follow-up were 29.7%, 29.3%, 39.3% and 51.3%

(p<0.001), much higher than after CABG (3.5%; p<0.001).

Completeness of revascularisation ___________________________________________________________________

34

Conclusion: Complete revascularisation by PTCA in multivessel coronary

disease did not result in a lower death or MI rate compared with incomplete

revascularisation. Overall the patient’s prognosis after PTCA is similar to

CABG, but patients with three or more remaining lesions after PTCA had a

worse prognosis than CABG patients.

Introduction

To date, percutaneous coronary interventions have a wide indication.

Whereas some years ago coronary bypass surgery was the dominant

therapeutic option in patients with multivessel coronary disease, due to

gained experience and the availability of sophisticated devices as

intracoronary stents, nowadays most of these patients are also eligible for

percutaneous treatment. Randomised clinical trials indicate that, at least in

the medium term, both revascularisation strategies are equally effective in

terms of mortality and the occurrence of myocardial infarction [1,2]. The re-

intervention rate, however, appears to be much higher after angioplasty,

especially during the first year. This can partly be explained by the

restenosis process. Another aspect is that complete revascularisation can

be difficult to realise by percutaneous intervention, especially in patients

with complex lesions. Therefore, interventional cardiologists regularly opt

for the strategy of incomplete revascularisation to relieve cardiac

symptoms. Recent reports indicated that this latter strategy is as safe and

effective as the strategy of complete revascularisation by percutaneous

treatment [3,7]. So far, however, no comparison has been made with

coronary artery bypass surgery. Furthermore, little is known about the

Chapter 3___________________________________________________________________

35

relationship between the degree of (in)completeness and the occurrence of

adverse events. The current paper concentrates on these two aspects, and

presents data from the Coronary Angioplasty versus Bypass

Revascularisation Investigation (CABRI) with respect to the completeness

of the revascularisation after coronary intervention and the occurrence of

cardiac complications at one-year follow-up.

Methods

Coronary Angioplasty versus Bypass Revascularisation Investigation

(CABRI)

CABRI is a multicentre, randomised trial comparing the strategies of

revascularisation by percutaneous transluminal coronary balloon

angioplasty (PTCA) and coronary artery bypass grafting (CABG) in patients

with multivessel coronary disease [8]. Included were patients <75 years of

age with a diameter stenosis (DS) <50% in at least two main epicardial or

large side branches as viewed online in two projections on the baseline

angiogram. At least one of the lesions had to be amenable to PTCA.

Patients with either stable or unstable angina pectoris were eligible, as well

as a-symptomatic patients who exhibited unequivocal ischaemia during

exercise testing. Excluded were patients with left main disease, overt

cardiac failure, those with a recent myocardial infarction (MI) or a recent

cerebrovascular accident, and patients with a previous PTCA or CABG.

The primary endpoint in CABRI was mortality and anginal status at one-

year follow-up. Secondary endpoints were MI and the need for repeat

revascularisation. CABRI was conducted in the first half of the 1990s, an


36

era in which neither stents nor glycoprotein IIb/IIIa inhibitors were available

in clinical practice.

Completeness of revascularisation

The location of lesions in the coronary tree is described according to the

American Heart Association 15-segment system (figure1) In patients

allocated to PTCA, an initial strategy of partial revascularisation was

allowed, i.e. not all significant lesions had to be attempted. Consequently,

for each lesion three possible situations could appear: no attempt was

made to dilate the lesion, a successful attempt was made, or an

unsuccessful attempt was made. A lesion was defined to be successfully

dilated if the post-procedural DS was less than 50%, as viewed online in

Figure 1: Schematic view of the coronary tree according to the American Heart Association 15-segments system. 1=proximal RCA; 2=mid RCA; 3=distal RCA; 4=postierior descending from dominant RCA; 5=left main stem; 6=proximal LAD (proximal to first septal); 7=mid LAD (distal to the first diagronal); 8=distal LAD (distal to the first diagonal); 9=first diagonal of LAD; 10=second diagonal of LAD; 11=main circumflex; 12=obtuse marginal (lateral) cirumflex; 13=mid circumflex; 14=posterolateral circumflex; 15=posterior descending from dominant LCA.

Chapter 3___________________________________________________________________

37

two angiographic projections. Revascularisation by percutaneous treatment

was defined as complete if all lesions were successfully dilated – i.e. if

there were no remaining lesions with DS <50% and incomplete

otherwise. A degree of (in)completeness was distinguished on the basis of

the number of remaining lesions. Surgery was viewed as the gold standard

for restoration of the coronary perfusion. Therefore, patients randomised to

PTCA who crossed to CABG (i.e. who underwent cardiac surgery before or

instead of a percutaneous intervention) were classified as completely

revascularised.

Data analysis

Patients randomised to PTCA were divided into separate groups according

to the completeness of the initial procedure. Patients who ultimately had a

complete revascularisation were distinguished from those with one, two,

and three or more remaining lesions, respectively. Differences in baseline

characteristics between these subgroups were evaluated by uncorrected 2-square test (dichotomous variables) ore one-way analyses of variance

(continuous variables), with statistical significance at the conventional

p=0.05 level.

The occurrence of major adverse cardiac events [death, MI, (re)CABG, and

(re)PTCA] during one-year follow-up is presented as a percentage.

Differences in event rates between patients randomised to CABG on the

one hand, and those randomised to PTCA with no or three or more

remaining lesions on the other hand, were evaluated by a series of two

uncorrected 2-square tests. To prevent increasing type I error, significance

of these tests was stated at p=0.025. Furthermore, within the group of

patients randomised to PTCA, the relation between the completeness of


38

revascularisation and the occurrence of adverse events was studied by

means of Mantel-Haenszel 2-square tests (significance at p=0.05).

Results

Angiographical success rates

In CABRI, 513 patients were randomised to CABG and 541 to PTCA.

Fifteen patients randomised to PTCA crossed to CABG, and the baseline

coronary angiogram was lacking in five additional patients. Thus there were

521 PTCA patients who were treated as randomised with complete pre-

and post-procedural angiographic information. Among these patients, 1856

significant lesions were reported on the baseline angiogram, implying an

average of 3.6 lesions per patient. There were 314 total occlusions (17% of

all lesions). A stenosis in the proximal part of the LAD was observed in 233

(13%) cases. An attempt to dilate a lesion was made in 66% of the non-

total occlusions (figure 2) and in only 39% of the total occlusions (p<0.001;

figure 3). The proximal LAD was more often attempted than the other

lesions (79% vs. 64%, p<0.001 in non-total occlusions and 57% vs. 38%,

p=0.121 in total occlusions). In the non-occluding lesions, 93% of the

attempts were successful. Although this percentage was significantly lower

in the total occlusions (p<0.001), 72% of the attempts in the latter lesions

were still successful.

Chapter 3___________________________________________________________________

39

Figure 2: Number of lesions in patients randomised to PTCA with a pre-procedural diameter stenosis >50% and <100%. The location of the lesions is indicated by the segment number according to the American Heart Association 15-segments system (see figure 1). The white, black and grey part of the bars indicate lesions that were attempted with success, attempted without success, and not attempted at all, respectively.

Figure 3: Number of lesions in patients randomised to PTCA with a pre-procedural diameter stenosis of 100% (total occlusion). For legends see figure 2.


40

Table 1: Baseline characteristics

Allocated strategy CABG PTCA Number of remaining lesions

Comparison of PTCA

subgroups0 1 2 3 (p-value)

Number of patients 513 148 147 122 119

Demographics and clinical presentationMedian age (25th, 75th percentile) (years)

61 (54,67)

60(53,66)

59(52,65)

61(57,66)

62(55,67)

0.145

Male gender (%) 78 76 84 68 81 0.018 Median LVEF (25th percentile, 75th percentile) (%)

65(55,73)

70(60,75)

65(56,73)

65(52,73)

63(53, 73)

0.100

Angina (CCS classification)No angina or class I/ class II or III/ class IV or unstable angina (%)

11/60/29 13/62/25 13/65/22 8/62/30 13/63/24 0.716

Function (NYHA classification)Class 0 or I/ class II/ class III or IV (%)

75/20/5 90/9/1 78/19/3 73/25/4 71/21/8 0.006

History Infarction (%) 42 40 42 33 53 0.020 Periph vessel disease (%) 6 6 4 8 11 0.188 CVA (%) 5 6 4 4 11 0.061 Smoking (%) 55 57 67 52 54 0.061 Diabetes mellitus (%) 12 9 9 14 17 0.142 Hypertension (%) 37 32 31 38 40 0.479

MedicationBeta-blockers (%) 65 67 64 61 50 0.091 Calcium antagonists (%) 65 65 66 69 71 0.784 Nitrates (%) 70 62 64 68 66 0.875

Lesion characteristics Number – 2/3/ 4/ 5 (%)

- 49/35/11/5

33/40/18/9

4/43/34/19

0/12/29/59

<0.001

Median (average) number - 3 (2.7) 3 (3.1) 4 (3.8) 5 (5.0) <0.001 Bifurcation (%) - 10 9 7 7 0.875 Eccentric (%) - 54 44 42 42 0.208 Long (%) - 73 83 80 96 <0.001 Tandem (%) - 12 7 4 8 0.207 Total occlusion (%) - 7 20 18 27 <0.001 Thrombus (%) - 3 1 2 4 0.391 Calcification (%) - 3 5 6 5 0.647

Chapter 3___________________________________________________________________

41

Completeness of revascularisation in patients randomised to PTCA

The initial intervention resulted in a complete revascularisation in 148 of the

patients randomised to PTCA (27%; 133 patient who were treated as

randomised plus 15 who crossed to CABG). In another 147 (27%) cases

there was only one remaining lesion, while there were 122 (23%) and 119

(22%) patients with two and three or more remaining lesions, respectively.

The five (1%) patients with a missing baseline angiogram could not be

classified. Table I presents the baseline characteristics of these groups.

There were some statistically significant differences in patient

demographics (gender), clinical presentation (NYHA class), and history

(previous infarction) between the four groups. Furthermore a significant

relationship was observed between the number of lesions at presentation

and the number of remaining lesions: e.g. 49% of patients with complete

revascularisation had two lesions at baseline, while this percentage was

only 4% among those with two remaining lesions. Also the length and

severity (total occlusion) of the lesion were significantly related to the

degree of completeness of revascularisation in patients randomised to

PTCA. On the other hand, the percentage occurrence of otherwise more

complex lesions (e.g. eccentric lesions, lesions with thrombus formation,

calcified lesions) was similar among the four groups.

Major adverse cardiac events

Table 2 presents the occurrence of cardiac complications in CABRI in

relation to the completeness of revascularisation. Follow-up information

was complete in all but four patients. Among the patients randomised to

PTCA, both short-term (up to 30 days after the initial procedure) and long-

term mortality was lowest in those with one remaining lesion after the initial


42

procedure (0.7% and 2.0%, respectively). Mortality was relatively high in

patients with complete revascularisation (2.1% at short term and 5.4% at

long-term follow-up) and in those with three or more remaining lesions

(2.5% and 5.0%). There was no significant relationship between the degree

of completeness of revascularisation and mortality, neither so if the group

of patients with no remaining lesions was ignored. The percentage

occurrence of MI (and the occurrence of either death or MI) showed a

similar pattern: lowest events rates in patients with one remaining lesion,

highest rates in those with no and three or more remaining lesions, and no

significant relation between the degree of completeness and short-term and

long-term event rates. The observed differences in mortality rates between

CABG and the PTCA sub-groups with no and three or more remaining

lesions were not statistically significant. The combined rate of death or MI

at one-year follow-up, however, was significantly higher in PTCA patients

with three or more remaining lesions than in those randomised to CABG

(12.6% vs. 6.2%; p=0.017).

The rate of subsequent coronary bypass operations among patients

randomised to PTCA was related to the degree of (in)completeness of the

initial procedure. During the first 30 days after this procedure, 2.8% of

patients with no remaining lesions underwent a CABG versus 20.2% of

those with three or more remaining lesions (p<0.001). At one-year follow-

up these percentages were 7.4% and 37.0% (p<0.001) respectively. In

contrast, the percentage occurrence of re-PTCA was similar among the

PTCA subgroups.

An interesting aspect is the difference in timing of repeat procedures

between the distinct subgroups. In patients with three or more remaining

lesions, 55 % of the CABGs (and 51% of the repeat interventions) were

Chapter 3___________________________________________________________________

43

performed within 30 days of the initial procedure. These percentages were

apparently lower in those with two and one remaining lesion: 41% and 15%

of the CABGs (and 33% and 14% of the repeat interventions) were done

within one month. In patients with a complete revascularisation, 38% of the

CABGs (and 24% of the repeat interventions) were performed during the

first month.

Discussion

Patients with multivessel coronary disease usually benefit from invasive

treatment strategies that aim at revascularisation of the coronary system:

coronary surgery or percutaneous intervention. To relieve patients’

complaints and restore myocardial perfusion it would seem logical to try to

achieve a complete revascularisation. However, several investigations have

demonstrated that a strategy of incomplete revascularisation is as safe (no

excess death or MI rate) and effective (pain relief) as complete

revascularisation [3-7].

The presented data confirm these conclusions to a large extent, although

some remarks need to be made.

Completeness of revascularisation and life threatening events

Among the PTCA patients who were treated as randomised, approximately

seven out of each ten non-occluding lesions were attempted with a success

rate of over 90%. Total occlusions were less frequently attempted, with a

lower, but still high rate of success. These data affirm other observations in

the pre-stent era [4]. A complete revascularisation was only realised in one


44

quarter of the patients, similar to findings in a registry of consecutive

patients with multivessel disease who underwent a PTCA in the same

period (17%) [3], and with results from the Multivessel Angioplasty

Prognosis Study (MAPS) [9].

The subdivision of patients randomised to PTCA according to the

completeness of the initial revascularisation procedure was not prespecified

by the protocol, and results of analyses on these subgroups should

therefore be interpreted with caution. Differences in event rates between

the PTCA subgroups might be biased by known or unknown differences in

baseline clinical and angiographical characteristics. Indeed, differences

were observed with respect to medical history and to the number and

severity of lesions to be treated. From a formal statistical point of view the

current data did not show a relation between the rate of death or

myocardial infarction during one-year follow-up and the degree of

completeness of revascularisation by PTCA. It is true that the percentage

occurrence of life-threatening complications moved upwards in patients

with one, two, and three or more remaining lesions, respectively, but event

rates were also high among patients with no remaining lesions. On the

other hand, PTCA patients with three or more remaining lesions had a

doubled risk of death or MI within the first year after the procedure

compared with bypass surgery.

Furthermore, the possibility of type II errors should not be ignored, as the

number of patients in each subgroup was relatively small. Finally, it cannot

be excluded that the relatively high event rate in patients with complete

revascularisation is caused by unjustifiable attempts to obtain an optimal

angiographic result.

Chapter 3___________________________________________________________________

45

Repeat interventions

Patients with remaining significant lesions after percutaneous intervention

most likely suffer from severe and diffuse coronary artery disease, and an

early bypass operation is therefore warranted. Indeed, the need for repeat

interventions to prevent life-threatening thrombotic events has been largely

recognised, as the percentage of CABG increased steeply with increasing

incompleteness of revascularisation by PTCA.

Patients with incomplete revascularisation not only underwent a repeat

intervention more often, but also much earlier than those with a more

complete revascularisation. Patients without remaining lesions formed an

exception: 38% of CABGs during follow-up were performed in the first

month. This observation might again be explained by a relatively high

percentage of ‘bail out’ situations after the PTCA procedure in this group of

patients, and leads as a matter of course to the most important limitation of

our investigation: CABRI was conducted in the pre-stent era. As the

immediate outcome of percutaneous treatment can be improved by

coronary stenting [10,11], the percentage of patients with a complete and

more stable result after percutaneous coronary intervention in today’s

clinical practice will likely be higher than in our population. Indeed, in the

Arterial Revascularisation Therapies Study (ARTS: a randomised

comparison of bypass surgery and coronary stenting in multivessel

coronary patients), 73% of the stent procedures resulted in a complete

revascularisation [12], which is considerably higher than the 27% obtained

after balloon angioplasty in our dataset. The favourable angiographic result

by coronary stenting, however, did not result in a improved outcome, as the

one-month death and MI rates in the angioplasty arms of CABRI and ARTS

were similar [12]. Hence, our data of the pre-stent era regarding the relation


46

between the completeness of revascularisation by percutaneous treatment

and patient’s prognosis are still valuable.

Percutaneous intervention of bypass surgery?

The strategy of complete revascularisation by PTCA in patients with

multivessel coronary artery disease did not result in a lower death or MI

rate during one-year follow-up as compared with a strategy of incomplete

revascularisation. Thus, complete revascularisation by percutaneous

intervention does not seem sensible in all patients. This is supported by the

observation that patients with one remaining lesion after PTCA had a lower

rate of life-threatening events than patients who initially underwent cardiac

surgery, be it at the cost of a higher re-intervention rate. Additional

analyses among patients with the worst angiographical result in patients

with multivessel coronary artery disease, coronary bypass surgery and

percutaneous intervention should not be played off against each other, but

should be considered as complementary parts of an umbrella coronary

revascularisation strategy. Treatment should be tailored to the individual

patient’s situation, taking into account the pros and cons of both strategies.

These considerations should also include cost aspects. In contrast to the

ARTS trial, however, CABRI was not designed to study cost-effectiveness

relations between percutaneous interventions and surgery. Future analyses

of the ARTS trial data may therefore be of great value in this respect.

Chapter 3___________________________________________________________________

47

References

1. Pocock SJ, Henderson RA, Rickards AF, et al. Meta-analysis of randomised trials comparing coronary angioplasty with bypass surgery. Lancet 1995;346:1184-9. 2. The Bypass Angioplasty Revascularisation Investigation( BARI) Investigators. Comparison of coronary bypass surgery with angioplasty in patients with multivessel disease. N Engl J Med 1996;335:217-25. 3. Bourassa MG, Hulubkov R, Yeh W, Detre KM. Strategy of complete revascularisation in patients with multivessel coronary artery disease (a report from the 1985-1986 NHLBI PTCA Registry). Am J Cardiol 1992;70:174-8 4. Bell MR. Bailey KR, Reeder GS, Lapeyre III, Holmes DR Jr Percutaneous transluminal angioplasty in patients with multivessel coronary disease: how important is complete revascularisation for cardiac event-free survival? J Am Coll Cardiol 1990;16:553-62. 5. Feyter PJ de, PTCA in patients with stable angina pectoris and multivessel disease; is complete revascularisation acceptable? Clin Cardiol 1992;15:317- 22. 6. Bell MR, Gersh BJ, Schaff HV, Holmes DR J, Fisher LD, Aldermann EL et al. Effect of completeness of revascularisation on long-term outcome of patients with three-vessel disease undergoing coronary artery bypass surgery: a report from the Coronary Artery Surgery Study (CASS) registry. Circulation 1992;86:446-57. 7. Bourassa MG, Yeh W, Holubkov R, Sopko G, Detre KM. Longterm outcome of patients with incomplete vs complete revascularisation after multivessel PTCA. A report from the NHLBI PTCA registry. Eur Heart J 1998;19:103-11. 8. The CABRI Trial Participants, First-year results of CABRI (Coronary Angioplasty versus Bypass revascularisation Investigation). Lancet 1995;346:1179-84. 9. Cowley MJ, Vandermael M, Topol EJ, Whitlow PL, Dean LS, Bulle TM, Ellis SG. Traditionally defined complete revascularisation needed for patients with multivessel disease treated by elective coronary angioplasty? Multivessel disease treated by elective coronary angioplasty? Multivessel Angioplasty Prognosis study (MAPS) Group. J Am Coll Cardiol 1993;22:1287-97. 10. Serruys PW, Jaegere P de, Kiemeneij F, Macaya, Rutsch W, Heyendrickx, et al. A Comparison of balloon expandable-stent implantation with balloon angioplasty in patients with coronary artery disease. Benestent study Group. N Engl J Med 1994;331:489-95. 11. Serruys PW, Hout B van, Bonnier H, Legrand V, Garcia E, Macaya C, et al. Randomised comparison of implantation of heparine-coated stents with balloon angioplasty in selected patients with coronary artery disease (Benestent II). Lancet 1998;352:673-81.


48

12. Brand MJ van den, Breeman A, Rensing A, et al. Comparison of angioplasty arms of two randomised trials of coronary angioplasty vesus surgery for multivessel coronary artery disease. Eur Heart J 1999;20:153(abstract P955).

CHAPTER 4

Characteristics, treatment and outcome of patients with non-ST-elevation acute coronary syndromes and multivesselcoronary artery disease: observations

from PURSUIT

A . Breeman, N. Mercado, M. Lenzen, M. van den Brand,

R.A. Harrington, R.M. Califf, E.J. Topol, M.L. Simoons, E. Boersma,

for the PURSUIT Investigators

Cardiology 2002;98:195–201

Chapter 4 ___________________________________________________________________

51

Abstract

Background: The 6-month clinical outcome of patients with multivessel

disease enrolled in PURSUIT (Platelet Glycoprotein IIb/IIIa in Unstable

Angina: Receptor Suppression Using Integrilin Therapy) is described.

Patients with complete angiography data were included; multivessel

disease was stratified according to the treatment strategy applied early

during hospitalization, i.e. medical treatment, percutaneous coronary

intervention (PCI) (balloon), PCI (stent), or coronary artery bypass grafting

(CABG).

Methods: Patients were divided into three groups according to the

treatment strategy applied during the first 30 days of enrolment. Patients

who did not undergo a percutaneous or surgical coronary intervention were

classified as medically treated. Patients who underwent a PCI (prior to a

possible CABG) were separated from those who underwent a CABG (prior

to a possible PCI). The PCI group was further subdivided: patients

receiving 1 coronary stents were separated from those in whom no stents

were used.

Results: The mortality rate at 30 days was 6.7, 3.9, 2.4 and 4.8% for the

medical treatment, PCI (balloon), PCI (stent) and CABG groups,

respectively (p value = 0.002). Differences as observed at 30 days were still

present at 6-month follow-up with 11.1, 5.8, 5.5 and 6.5% mortality event

rates for the aforementioned groups (p value = 0.002). The 30-day

myocardial infarction (MI) rate according to the opinion of the Clinical

Events Committee was lower among medically than non-medically treated

patients, with the highest event rate observed in the CABG group (27.7%).

Multivessel disease in NSTE-ACS ___________________________________________________________________

52

Approximately half of the MIs in the PCI and CABG subgroups occurred

within 48 h after the procedure.

Conclusions: The observed differences in clinical outcomes are explained

by an imbalance in baseline characteristics and comorbid conditions

between the analyzed groups of patients.

Introduction

Several clinical trials have been performed to evaluate whether patients

with coronary artery disease benefit most from medical treatment only,

percutaneous coronary intervention (PCI), or coronary bypass surgery

(CABG) [1,2]. Other studies have specifically compared percutaneous

transluminal coronary balloon angioplasty (PTCA) against CABG [3–9] and

finally contemporary trials of coronary stenting and optimal adjunctive

pharmacological therapy versus CABG have recently been reported [10–

12]. However, most of these studies have predominantly included patients

with chronic stable angina and few data are available on the characteristics

and clinical outcome of patients with multivessel disease presenting with an

acute coronary syndrome.

The Platelet Glycoprotein IIb/IIIa in Unstable Angina: Receptor Suppression

Using Integrilin Therapy (PURSUIT) trial was a large-scale randomized

clinical trial on the effects of eptifibatide versus placebo in patients with

acute coronary syndromes without persistent ST elevation [13]. As the

enrolment criteria were broad, PURSUIT encompasses a wide variety of

patients, hospital settings and treatment policies, and therefore accurately

reflects standard clinical practice.

Chapter 4 ___________________________________________________________________

53

The aim of this study was to describe the characteristics and short-term

clinical outcome of patients with multivessel coronary artery disease in the

PURSUIT population according to the treatment strategy applied early

during hospitalization.

Materials and Methods

Patient population

The design and methods of the PURSUIT trial have been previously

published [13]. In summary, patients were eligible if they presented within

24 h of an episode of ischemic chest pain (>10 min), and had either

transient ST elevation (>0.5 mm), transient or persistent ST depression (0.5

mm), T wave inversion (>0.1 mm), or elevation of the creatine kinase MB

fraction (CK-MB) above the upper limit of normal (ULN). Patients with

persistent (>30 min) ST elevation were excluded. Eligible patients were

randomly assigned to treatment with eptifibatide or placebo. All other

treatment decisions, including the use and timing of PCI or CABG were left

at the discretion of the treating physician.

Coronary angiography was performed within 30 days of enrolment in 5,937

(63%) of the 9,461 patients who participated in PURSUIT (figure 1). Among

patients with complete angiographic data, 3,067 (58%) had a significant

stenosis (>50% diameter stenosis by visual inspection) in 2 major native

coronary arteries or in the left main stem. These patients were classified as

having multivessel coronary artery disease and are the subjects of interest

for the current analysis.


54

Classification according to applied treatment strategy Patients were divided into three groups according to the applied treatment

strategy during the first 30 days of enrolment. Patients who did not undergo

a percutaneous or surgical coronary intervention were classified as

medically treated. Patients who underwent a PCI (prior to a possible

CABG) were separated from those who underwent a CABG (prior to a

possible PCI). The PCI group was further subdivided: patients receiving 1

Figure 1: PURSUIT patient population flow chart stratified accoording tot the number of diseased vessels and treatment strategy applied to patients with multivessel coronary disease. VD = vessel disease.

Chapter 4 ___________________________________________________________________

55

coronary stents were separated from those in whom no stents were used

(figure 1).

Definition of myocardial infarction The primary endpoint of PURSUIT was a composite of death or non-fatal

myocardial infarction (MI) at 30 days. A computerized algorithm was used

to review the clinical events. If a possible event was identified, further

documentation was collected and the case reviewed in detail and

adjudicated by a central Clinical Events Committee (CEC). MI was

diagnosed on the basis of new ST segment elevations, new Q waves, or

new or repeated CK-MB elevations above the ULN. Following

percutaneous or surgical intervention, the elevation of CK-MB level was

required to be at least 3–5 times the ULN.

Data analysis The statistical analysis was performed using the SAS 8.0 software package

(SAS Institute, Cary, NC, USA). Continuous variables are presented as

mean values±standard deviation and dichotomous variables as

percentages. One-way analyses of variance (ANOVA) and X2-tests were

applied to evaluate differences in baseline characteristics, death, MI rates

and rates of repeated coronary interventions (both at the 30-day and 6-

month follow-up) between the different subgroups classified according to

the applied treatment strategy. In case of a statistically significant difference

in clinical events or repeated coronary interventions, which was specified at

the conventional p<0.05 level, repeated tests were performed to further


56

evaluate the inter-subgroup differences. The Bonferroni method [14] was

used to avoid spurious significant results after multiple testing, and the level

of significance was lowered to p<0.0167 and p<0.0083 in case of 3 (balloon

– stent – surgery) and 6 possible comparisons (medical treatment – balloon

– stent – surgery), respectively.

Results

Patient characteristics During the first 30 days of randomization, a PCI was performed in 1,075 of

3,067 (35%) patients, with stent placement in 542 cases (50% of the PCI

procedures), whereas 1,091 of 3,067 (36%) patients underwent CABG. The

remaining 901 (29%) patients were medically treated (figure 1). A

significant difference was evident between the treatment subgroups with

respect to age; medically treated and CABG patients were older than those

undergoing PCI (table 1). There were also differences regarding history of

prior cardiovascular events and interventions. Almost one quarter of the

PCI patients had a previous PCI versus 17 and 14% in the medical and

surgical subgroups, respectively. A prior CABG was performed in 36% of

medically treated patients, and this figure was only 9% in the CABG

subgroup. A history of MI, heart failure, cerebrovascular accident, as well

as peripheral vessel disease was more frequently observed in the medically

versus non-medically treated patients. No important differences were

observed in the use of cardiac medications except for the use of ACE

inhibitors, which was more frequent in medically treated patients.

Chapter 4 ___________________________________________________________________

57

Table 1: Clinical baseline characteristics.

Medical PCI (n=1,075) CABG

(n=901) balloon(n=533)

stent(n=542) (n=1.091)

Demographics Mean age ± SD, years 65 ± 10 63 ± 11 62 ± 11 64 ± 10*** Male gender, % 70 74 73 73 Caucasian, % 86 89 91 90*

Medical history and risk factors

Hypertension, % 62 60 61 57 Diabetes mellitus, % 30 26 23 27 Current smoker, % 28 29 29 26 Hypercholesterolemia, % 50 46 53 47 Prior PCI, % 17 23 23 14*** Prior CABG, % 36 24 29 9*** Prior MI, % 47 40 36 34*** Heart failure, % 14 8 10 6*** Prior CVA,% 7 4 5 3** Peripheral vessel disease, % 13 8 9 9**

Cardiac medication prior to admission

Aspirin, % 73 74 76 74 Beta-blocker, % 48 50 49 49 Calcium antagonist, % 39 34 36 36 Nitrates, % 73 75 77 74 ACE inhibitors, % 30 23 24 21***

Clinical presentation CK-MB >1 ULN, % 52 52 44 48* ST depression >0.5 mm, % 52 47 52 58*** ST elevation >0.5 mm, % 12 17 15 13* T wave inversion >0.5 mm, % 48 51 50 45

ACE = Angiotensin converting enzyme; CVA = cerebrovasculair accident; other abbreviations, as defined in the text. Statistical test (integral comparison of 4 groups); *p<0.05; **p< 0.01; ***p<0.001.


58

Table 2: Coronary angiography results.

Medical PCI CABG

balloon stent Vessel disease, % 2 38** 62** 61** 28*** 3 51 32 30 52 LM 11 6 9 21

Significant stenosis (DS > 50%) in RCA, % 84 79 81 85** LAD, % 88 77 82 91*** LCX, % 84 77 72 79*** LM, % 12 6 10 22***

Total occlusion (DS = 100 %) in RCA, % 48 33 33 33*** LAD, % 35 24 26 20*** LCX, % 32 21 21 17***

Culprit artery, % RCA 14 27 25 16*** LAD 21 25 29 29*** LCX 14 31 21 9 LM 2 1 1 9*** Graft 9 8 14 3*** None/unknown 40 8 10 34*** Mean LVEF± SD 50 ±16 55 ±14 54 ± 14 53 ± 14***

DS = Diameter of stenosis; LAD = left artery, descending; LCX = left circumflex; LM = left main; LVEF = left ventricular ejection fraction; RCA = right coronary artery; other abbreviation, as defined in the text. Statistical test (integral comparison of 4 groups): **p<0.01; ***p<0.001.

Angiographic findings

Patients who underwent CABG had more severe coronary artery disease

(52% had 3-vessel and 21% left main disease), immediately followed by

medically treated patients with a similar percentage of 3-vessel disease

(51%) but less often, left main disease (11%) (table 2). A total occlusion in

any of the major native coronary arteries was more often present in

Chapter 4 ___________________________________________________________________

59

medically than in non-medically treated patients. Medically and surgically

treated patients not only had more severe, but also more diffuse coronary

artery disease than PCI patients, as in 34% (CABG) to 40% (medically

treated) of the patients, the culprit artery could not be identified; this

percentage was only 8–10% in PCI patients. Left ventricular ejection

fraction was lowest among medically treated patients. There were no

apparent differences in coronary angiography results between PCI patients

receiving stents and those that did not receive stents.

Clinical outcome

The 30-day mortality rate was significantly higher among medically treated

patients (6.7%) than among those undergoing PCI either with (2.4%) or

without stent placement (3.9%) (figure 2). The observed difference in

mortality rate (p value = 0.067) between medically treated and CABG

patients (4.8% mortality) did not reach the required level of significance,

which was prespecified as p<0.0083.

The 30-day MI rate according to the opinion of the CEC was lower among

medically than non-medically treated patients, with the highest event rate

observed in the CABG group (27.7%); approximately half of the MIs in the

PCI and CABG subgroups occurred within 48 h after the procedure.

Differences in event rates as observed at 30 days were still present at the

6-month follow-up. Mortality was highest in the medically treated subgroup

(11.1%). Mortality rates were similar in the non-medical treatment

subgroups (ranging from 5.5 to 6.5%). MI rates as judged by the CEC were

lowest in the medically treated patients (20.8%) and highest in patients

undergoing CABG (29.6%).


60

Repeat Revascularization Procedures

Figure 2: 30-day (a) and 6-month (b) clinical outcome according to the treatment strategy applied. Black bars indicate mortality and white bars indicate MI adjudicated by the CEC.p values for the overall comparison between any of the treatment strategies applied (medical treatment, balloon, stent or surgery) and each clinical endpoint are as follows: adeath: p=0.002, MI: p=0.001 both at 30 days and 6 months.

Figure 3: 30-day (a) and 6-month (b) repeat revascularization procedures according to the treatment strategy applied. Black bars indicate PCI and white bars indicate CABG. p values for the overall comparison between any of the re-interventions and each group based on the treatment strategy applied initially (medical treatment, balloon, stent or surgery) are as follows: PCI, CABG: p<0.001 both at 30 days and 6 months.

Chapter 4 ___________________________________________________________________

61

The rate of repeat revascularization at the 30-day and 6-month follow-up

were significantly lower after CABG than after PCI (figure 3). Patients

undergoing stent implantation during the initial PCI had lower CABG rates

at each of these 2 points in time when compared to non-stented patients.

No apparent differences were observed in the rates of repeat interventions

between stented and non-stented patients. A substantial number of

medically treated patients still underwent a PCI (4.6%) or CABG (15.4%)

procedure between 1 and 6 months after admission.

Discussion

Patients who present with acute chest pain without persistent ST segment

elevation represent a heterogeneous population, which spans from non-

cardiac chest pain (retrospectively diagnosed), to unstable angina and

acute MI. The uncertainty in early clinical diagnosis forces clinicians to

embark upon an empirical course of treatment, and this is the main reason

why the clinical community still debates intensively regarding the optimal

treatment strategy for patients with non-ST elevation acute coronary

syndromes. Coronary angiography identifies patients with non-significant

coronary stenoses and those with multivessel or left main disease. The

former group has an excellent prognosis with a low risk of progression to MI

or death, whereas the latter group, which is at an increased risk of

progression to any of the aforementioned events, may derive a survival

benefit from revascularization (either PCI or CABG) [15,16]. Patients who

are not suitable candidates for standard revascularization or those who are

at high risk of major peri-operative complications due to comorbid


62

conditions represent a distinct category in which medical treatment is

preferred.

A major goal in PURSUIT was to understand the heterogeneity of the

patient population and treatment strategies applied. The investigators

therefore chose to embed the study of the effects of epifibatide in a real-life

clinical setting including a broad spectrum of clinical practices, from rural

hospitals to major tertiary referral centers around the world. To reflect

actual clinical practice, no recommendations were made regarding the use

and timing of coronary angiography, percutaneous coronary interventions

or coronary bypass surgery, but all treatment decisions were left at the

discretion of the team of treating physicians. Therefore, the results of the

present descriptive analysis should be interpreted with this background in

mind and viewed with the inherent limitations to subgroup analysis of

randomized clinical trials [17].

Although not prospectively randomized to each of the treatment strategies

compared, it is important to note that the medical therapy, early PCI and

CABG ratio in these subgroups of 3,067 patients with an acute coronary

syndrome and multivessel coronary artery disease was almost 1:1:1.

Indeed, important differences were observed in clinical characteristics and

coronary anatomy between the distinct subgroups. Patients who did not

undergo a coronary intervention within 30 days after enrolment, generally

were in a less favourable clinical condition than patients undergoing early

invasive treatment. The relatively high 30-day and 6-month mortality rate

among medically treated patients is therefore not surprising and argues for

the search of better treatment strategies in unstable patients with

multivessel coronary artery disease that are not good candidates for

revascularization procedures. Important determinants in the decision to

Chapter 4 ___________________________________________________________________

63

refrain from invasive treatment in this patient population seem to be

comorbid conditions, left ventricular dysfunction (medically treated patients

more often had a history of CABG, heart failure, and a worse left ventricular

function as compared to CABG patients) and the extent of coronary artery

disease (medically treated patients more often had 3-vessel and left main

disease as compared to PCI patients).

Limitations

This was a retrospective assessment of clinical, angiographic

characteristics and clinical events in patients enrolled in a multicenter

clinical trial and stratified according to the treatment strategy applied with a

follow-up limited to 6 months, which can be considered as the main caveat

of this study. We lack data on anginal status at baseline and 6 months; and

on other predictors of adverse outcome such as completeness of

revascularization; in both PCI and CABG patients. There were insufficient

data on post-procedural cardiac enzymes as well. However, the present

analysis reflected standard practice in a wide range of clinical settings, and

contemporary treatment strategies for the management of patients with

acute coronary syndromes and multivesel disease were used in this trial.

Conclusions

The observed major differences in clinical outcome are explained by an

imbalance in baseline and angiographic characteristics between the groups

of patients analyzed according to the treatment strategy applied.


64

References

1. Hueb WA, Bellotti G, de Oliveira SA, Arie S, de Albuquerque CP, Jatene AD, Pileggi F. The Medicine, Angioplasty or Surgery Study (MASS): A prospective, randomized trial of medical therapy, balloon angioplasty or bypass surgery for single proximal left anterior descending artery stenoses. J Am Coll Cardiol 1995;26:1600–1605. 2. DiMarco J. Results from late-breaking clinical trials sessions at ACC 2001. J Am Coll Cardiol 2001;38:595–612. 3. Puel J, Karouny E, Marco F, Asoun B, Galinier M, Elbaz M, Alibelli MJ, Bounhoure JP. Angioplasty versus surgery in multivessel disease: Immediate results and in-hospital outcome in a randomized prospective study. Circulation 1992;86(Suppl I):372.4. Coronary angioplasty versus coronary artery bypass surgery: the Randomized Intervention Treatment of Angina (RITA) trial. Lancet 1993;341:573-580. 5. Rodriguez A, Boullon F, Perez-Balino N, Paviotti C, Liprandi MI, Palacios IF. Argentine randomized trial of percutaneous transluminal coronary angioplasty versus coronary artery bypass surgery in multivessel disease (ERACI): In- hospital results and 1-year follow-up. ERACI Group. J Am Coll Cardiol 1993;22:1060–7. 6. Hamm CW, Reimers J, Ischinger T, Rupprecht HJ, Berger J, Bleifeld W. A randomized study of coronary angioplasty compared with bypass surgery in patients with symptomatic multivessel coronary disease. German Angioplasty Bypass Surgery Investigation (GABI). N Engl J Med 1994;331:1037–1043. 7. King SB 3rd, Lembo NJ, Weintraub WS, Kosinski AS, Barnhart HX, Kutner MH, Alazraki NP, Guyton RA, Zhao XQ. A randomized trial comparing coronary angioplasty with coronary bypass surgery. Emory Angioplasty versus Surgery Trial (EAST). N Engl J Med 1994;331:1044–50. 8. First-year results of CABRI (Coronary Angioplasty versus Bypass Revasculari- sation Investigation). CABRI Trial Participants. Lancet 1995;346:1179–1184. 9. Comparison of coronary bypass surgery with angioplasty in patients with multivessel disease. The Bypass Angioplasty Revascularization Investigation (BARI) Investigators. N Engl J Med 1996;335:217–225. 10. Rodriguez A, Bernardi V, Navia J, Baldi J, Grinfeld L, Martinez J, Vogel D, Grinfeld R, Delacasa A, Garrido M, Oliveri R, Mele E, Palacios I, O’Neill W. Argentine Randomized Study: Coronary Angioplasty with Stenting versus Coronary Bypass Surgery in patients with Multiple-Vessel Disease (ERACI II): 30-day and one-year follow-up results. ERACI II Investigators. J Am Coll Cardiol 2001;37:51–58. 11. Serruys PW, Unger F, Sousa JE, Jatene A, Bonnier HJ, Schonberger JP, Buller N, Bonser R, van den Brand MJ, van Herwerden LA, Morel MA, van Hout BA. Comparison of coronaryartery bypass surgery and stenting for the treatment of multivessel disease. N Engl J Med 2001;344:1117–1124.

Chapter 4 ___________________________________________________________________

65

12. Stables RH. Coronary artery bypass surgery versus percutaneous coronary intervention with stent implantation in patients with multivessel coronary artery disease (the Stent or Surgery trial): A randomized controlled trial. Lancet 2002;360:965–970. 13. Inhibition of platelet glycoprotein IIb/IIIa with eptifibatide in patients with acute coronary syndromes. The PURSUIT Trial Investigators. Platelet Glycoprotein IIb/IIIa in Unstable Angina: Receptor Suppression Using Integrilin Therapy. N Engl J Med 1998;339:436–443. 14. Bland JM, Altman DG. Multiple significance tests: the Bonferroni method. BMJ 1995;310:170. 15. Bertrand ME, Simoons ML, Fox KA, Wallentin LC, Hamm CW, McFadden E, de Feyter PJ, Specchia G, Ruzyllo W. Management of acute coronary syndromes: Acute coronary syndromes without persistent ST segment elevation; recommendations of the Task Force of the European Society of Cardiology. Eur Heart J 2000;21:1406–1432. 16. Braunwald E, Antman EM, Beasley JW, Califf RM, Cheitlin MD, Hochman JS, Jones RH, Kereiakes D, Kupersmith J, Levin TN, Pepine CJ, Schaeffer JW, Smith EE, 3rd, Steward DE, Theroux P, Gibbons RJ, Alpert JS, Eagle KA, Faxon DP, Fuster V, Gardner TJ, Gregoratos G, Russell RO, Smith SC, Jr. ACC/AHA guidelines for the management of patients with unstable angina and non-ST-segment elevation myocardial infarction: Executive summary and recommendations. A report of the American College of Cardiology/American Heart Association task force on practice guidelines (committee on the management of patients with unstable angina). Circulation 2000;102:1193– 1209. 17. Yusuf S, Wittes J, Probstfield J, Tyroler HA. Analysis and interpretation of treatment effects in subgroups of patients in randomized clinical trials. JAMA 1991;266:93–98.

CHAPTER 5

Long-term follow-up after invasiveapproach of coronary artery disease

in daily practice

A. Breeman, J. Timmer, J.P. Ottervanger, E. Kolkman

E. de Kluiver, H. Rigter, P. Boonstra, F. Zijlstra

Int J Cardiol 2005;105(2):186-91

Chapter 5 ___________________________________________________________________

69

Abstract

Objectives: To assess long-term survival in unselected patients with

coronary artery disease in who an invasive approach is considered.

Methods: All patients with significant coronary artery disease who were

presented for coronary revascularisation to two tertiary centres in 1992

were included. Follow-up data were collected in September 2002.

Multivariate Cox’ proportional-hazards regression analysis was applied to

assess the independent relation between variables and 10-year survival.

Results: A total of 877 patients were included in this analysis. Mean age

was 62 and the most common clinical diagnosis was chronic stable angina

(60%). Diabetes was present in 12% of the patients. During the follow-up

period, a total of 233 patients (27%) died. Predictors of long-term survival

were increasing age, diabetes, peripheral vascular disease and a

decreased left ventricular function. Compared to medical treated patients,

those treated with revascularisation (either by PCI or CABG) had a

decreased long-term mortality (p<0.05). Of the patients with PCI 27% had

died, compared to 24% in those who had CABG and 36% of those who

were treated medically. However, after adjusting for differences in baseline

variables, conservative treatment was no significant predictor of long-term

mortality. After multivariable analyses, increasing age, decreased left

ventricular function and diabetes were independent predictors of long-term

mortality.

Conclusions: In patients with coronary artery disease in whom an invasive

approach is considered, increasing age, impaired left ventricular function

and diabetes are the strongest predictors of long-term mortality. After

Long-term results after invasive approach ___________________________________________________________________

70

adjustments for differences in baseline variables, invasive treatment is not

associated with a lower long-term mortality.

Introduction

Several medical treatments of coronary artery disease reduce mortality [1-

4] . Long-term results after an invasive approach, however, are conflicting

and may differ between Coronary Artery Bypass Surgery (CABG) and

Percutaneous Coronary Intervention (PCI) [5-8]. In a meta-analysis of all

randomised controlled trials comparing CABG with PCI, CABG was

associated with lower five-year mortality, less angina, and fewer recurrent

revascularisation procedures [9]. However, most of these trials had

important limitations. Approximately two-thirds of the eligible patients were

excluded because of angiographic finding, such as chronic total occlusions,

complex stenoses, left main coronary artery disease, or the inability to

achieve adequate revascularisation. Patients with recent myocardial

infarction or previous revascularisation were often excluded as well. So,

these results concerned selected patients and may be not comparable to

results observed in daily practice. Data from registries or a consecutive

patient population may provide more insights in long-term outcome after

coronary revascularisation. However, many data from registries, such as

from the Coronary Artery Surgery Study (CASS) Registry are of older date

and may not reflect the present state of the art [10,11].

To evaluate long-term prognosis of patients with coronary artery disease in

daily practice, we did a long-term follow-up study of patients included in the

DUCAT (DUtch inventory of invasive Coronary Atherosclerosis Treatments)

Chapter 5 ___________________________________________________________________

71

study. In particular, we assessed the association between type of

revascularisation (CABG or PCI) and mortality.

Methods

The DUCAT study aimed to determine the appropriateness of decisions to

offer patients with coronary artery invasive or medical treatment [12]. The

appropriateness of decisions was assessed using the RAND/UCLA method

[13]. The DUCAT expert panel consisted of six cardio-thoracic surgeons

and 6 interventional cardiologists of the 12 heart centers in the

Netherlands. This panel method proved to be consistent and reliable in

assessing appropriateness.

Enrolment began in February 1992 for a period of 3 months. Each case

was presented by the patient’s cardiologist in person or by letter, fax, or

telephone, and eventually led to an intention to treat decision in favour of

PCI, CABG, or medical treatment. All presentations were based on

coronary angiographic results, showing significant coronary artery disease,

defined by the DUCAT panel as at least 50% narrowing of the left main

artery or 70% in other arteries. Patients who previously had had CABG or

for whom CABG was to be combined with other surgery (cardiac or

general) were excluded. At the time of presentation to the heart team, key

patient data were collected, including demographics, medical history, risk

factors for coronary artery disease, symptomatology, medication, ischemia

detection tests, coronary angiographic evaluation, left ventricular function

tests, urgency status and intention of treatment. The main findings of the

DUCAT study have been published previously [12,13]. In summary, 3646


72

consecutive patients were included. Unstable angina was the clinical

condition associated with the highest proportion of appropriate intention-to-

treat decision, whereas in asymptomatic coronary disease the rate of

appropriate decisions was lowest.

The present study is a two-center follow-up study of all patients

consecutively presented to the heart centers of Zwolle and Groningen in

the Netherlands. In these two centers, a total of 1047 patients were

presented to the cardio-pulmonal teams.

Follow-up data were obtained on the vital status up to ten years after

inclusion. Follow-up data were collected via the register office, the General

Practitioner or via a direct contact with the patient or his relatives by

telephone.

Statistics

Differences between group means were tested by two-tailed Student's

t-test. A Chi-square statistic was calculated to test differences between

proportions. Survival functions were calculated, using the Kaplan-Meier

product limit method [14]. Mantel-Cox (or log-rank) test was applied to

evaluate the differences between survival functions [15].


describe the independently relation between variables and 10-year survival.

Statistical significance was defined as a p-value of less than 0.05. The

Statistical Package for the Social Sciences (SPSS Inc, Chicago, IL, USA)

version 11.0 was used for all statistical analyses.

Chapter 5 ___________________________________________________________________

73

Results

General characteristics

During enrolment, a total of 1047 patients were presented to the two heart

centers for the first time. Of these patients, 877 patients (84%) met the

inclusion criteria and were enrolled in the study. Mean age of the patients

was 62.4 years (SD 9.6). Females were older than males, 66 years vs. 61

years respectively (p<0.01). The most common clinical diagnosis was

chronic stable angina (60%). Anti-ischemic medication was used by 94% of

the patients. A history of myocardial infarction was reported in 37% and

29% had hypertension. Diabetes was present in 12% of the patients.

Table 1: Baseline characteristics of 877 patients according to revascularisation type.

Characteristic PCI CABG Medical therapy

No 376 384 117 Mean age (yr) 61 10 63 9* 63 11 Males 263 (69.9%) 299 (77.9%)# 90 (76.9%) Diabetes 39 (10.5%) 51 (13.3%) 17 (14.7%) Previous MI 164 (43.9%) 194 (50.9%)# 69 (59.0%)* Previous PCI 52 (13.9%) 33 (8.6%) 11 (9.4%) COPD * 30 (8%) 36 (9.4%) 12 (10.3%) Hypertension 117 (31.4%) 114 (29.8%) 27 (23.1%) Peripheral vasc. disease 29 (7.8%) 39 (10.2%) 19 (16.2%)#

Smoking 114 (33.1%) 81 (23.3%)# 31 (29.2%) Hypercholesterolaemia 117 (33.9%) 149 (41.9%) 30 (29.7%) Single vessel disease 231 (61%) 17 ( 4.4%)* 58 (50%)#

Left main stenosis 3 (0.8%) 55 (14%)* 3 (0.8%) Poor LV function 9 (2.5%) 5 (1.3%) 11 (9.5%)* Unstable 153 (41.2%) 118 (31.4%)* 15 (3.2%)*CCS III, IV 118 (31.8%) 141 (37.5%) 51 (44.7%)#

CCS I, II, (Atyp, Asym) 77 (20.8%) 115 (30.6%)* 47 (41.2%)* Acute MI 23 (6.2%) 2 (0.5%)* 1 (0.9%)#

Type C lesion 109 (29%) 259 (67.4%)* 75 (64.1%)*Prox Lad lesion 131 (34.8%) 229 (59.6%)* 40 (34.2%)

Difference with ‘PCI’ significant at p<0.05 (#) or p<0.01 (*). CVA = Cerebro Vasculair Accident; COPD = Chronic Obstructive Pulmonary disease. CCS = Canadian Cardiovascular Society.


74

Angiography and revascularisation

Of the included patients, 30% had three-vessel disease. Left ventricular

(LV) function was normal in 81%, whereas 16% had moderately decreased

LV function and 3% had a severely depressed LV function.

For 376 patients (43%) the cardiopulmonal team preferred PCI as

appropriated for 384 patients (44%) CABG and for 117 patients (13%)

medical therapy. Table 1 shows the differences in baseline characteristics

between the three treatment groups. Patients treated with PCI were

younger, were more often female, and had more frequently unstable

coronary artery disease, less extensive coronary artery disease. Medical

therapy was more often preferred for patients with complex lesions,

previous myocardial infarction, peripheral vessel disease and poor left

ventricular function. As expected, patients treated with CABG had more

extensive coronary artery disease.

Long-term mortality

During the 10-years follow-up period, a total of 233 patients (27%) died.

Table 2 presents the association between baseline characteristics and

long-term mortality. In these univariate analyses, increasing age, unstable

coronary artery syndrome, diabetes, peripheral vascular disease, and poor

left ventricular function at baseline were statistically significant predictors of

long-term mortality. Patients scheduled for medical treatment had a higher

long-term mortality than patients scheduled for an invasive approach. Of

the patients for whom PCI was recommended, 27% had died during follow-

up, compared to 24% for CABG and 36% for medical treatment (p<0.05).

Survival of the patients according PCI, CABG or medical treatment is

depicted in figure 1.

Chapter 5 ___________________________________________________________________

75

Table 2: Predictors of long-term mortality in univariate analysis

Variable Odds Ratio 95% confidence interval Male gender 0.907 0.64 - 1.27 Revascularisation 0.599 0.40 - 0.91 Age (year) 1.103 1.08 - 1.13 Previous MI 1.096 0.82 - 1.48 Hypertension 1.141 0.82 - 1.58 Diabetes mellitus 1.724 1.13 - 2.64 Periph. vasc. disease 2.026 1.28 - 3.20 Smoking 0.986 0.69 - 1.41 COPD 1.260 0.76 - 2.09 Single vessel disease 0.992 0.73 - 1.36 Left main stenosis 1.128 0.63 - 2.02 Poor LV function 3.229 1.45 - 7.19 Unstable angina 0.963 0.70 - 1.33 CCS III, IV 1.328 0.97 - 1.81 CCS I, II, atypical 0.671 0.47 - 0.96 Acute MI 1.747 0.78 - 3.91 Pos concl stress test 2.009 1.09 - 3.71 Prox Lad 0.883 0.65 - 1.19 Type C lesion 1.106 0.82 - 1.49 Nitrates 0.903 0.65 - 1.26 Betablockers 0.909 0.67 - 1.24

Figure 1: Long-term mortality of 877 patients according to medical treatment, CABG or PCI.

Medical compared to PCI: P value = 0.0201

CABG compared to PCI: P value = 0.5180

Medical compared to CABG: P value = 0.0268


76

Table 3: Predictors of long-term mortality after multivariate analysis.

Variable Odds Ratio 95% confidence interval Male gender 1.36 0.98 - 1.89 Age (per year) 1.09 1.07 - 1.12 Conservative treatment# 1.24 0.86 - 1.79 Decreased LV function 1.58 1.15 - 2.18 Diabetes 1.46 1.01 - 2.13 Periph. Vasc. Disease 1.28 0.85 - 1.91

#As compared to revascularisation by either PCI or CABG. Adjusted for differences in the other variables.

The predictors of long-term mortality differed between the groups of

patients treated conservative, by CABG or by PCI. Admission with an acute

coronary syndrome or decreased left ventricular function (OR 1.73; 95% CI

1.09-2.29) were predictors of higher mortality in patients scheduled for

CABG. Diabetes was particularly a predictor of mortality in patients

scheduled for PCI or medical treatment (OR 2.33; 95% CI 1.51-3.59).

Patients with hypertension or unstable angina had a higher mortality after

medical treatment. Peripheral vascular disease predicted mortality in

patients scheduled for PCI more strongly than in patients treated by CABG

of medically.

Multivariable analyses

To adjust for potential confounders, multivariable analyses were performed.

Adjustments were made for age, gender and statistically significant

predictors of long-term mortality in the univariate analyses. In the

multivariable model, age, gender, revascularisation, diabetes, peripheral

vascular disease and decreased left ventricular function were included. The

results of multivariable analyses are summarised in table 3. Increasing age,

Chapter 5 ___________________________________________________________________

77

impaired left ventricular function and diabetes were independent, significant

predictors of long-term mortality. Revascularisation was not statistically

significant associated with a lower mortality.

Discussion

In this retrospective analysis of unselected patients with coronary artery

disease presented for an invasive approach, independent predictors of

long-term mortality were increasing age, impaired left ventricular function

and diabetes. After adjustments for differences in baseline variables,

invasive treatment by either PCI or CABG was not associated with a lower

long-term mortality.

Although randomised clinical trials are the gold standard for evidence

based medicine, extrapolation of the results to the general patient

population should be done with caution. In most randomised trials, many

patients are excluded and crossover between treatment groups is common.

Therefore, data from cohort studies concerning patients seen in daily

practice remain an important source of information.

In our study, patients with conservative treatment had not a significant

increased risk of long-term mortality after multivariable analyses. These

findings are in line with the results of several previous studies. In the latest

meta-analysis of all randomised studies comparing CABG and PCI, in

patients with single vessel disease there was no significant survival benefit

after CABG after eight years follow-up [9]. Also in the RITA-II trial the 7-

years survival curves of patients randomised to PCI or medical treatment

overlapped [16]. Furthermore, the meta-analysis including the VA,


78

European Co en CASS showed after a 10-year follow-up period

comparable mortality for patients randomised to CABG. Also, the results of

revascularisation in patients with an acute coronary syndrome are not clear.

In the OASIS registry, evaluating approximately 8000 patients with

suspected unstable angina or myocardial infarction without initial ST

elevation from 6 countries, after revascularisation a reduction in refractory

ischemia and need for hospitalisation was observed, but not a decrease of

mortality [17].

The lack of survival benefit of revascularisation in our analysis may have

several explanations. Revascularisation may be beneficial only after short

or intermediate follow-up, whereas after long-term follow-up mortality is

comparable between patients with and without revascularisation. Another

explanation is that patients allocated to invasive treatment have more

unfavourable prognostic characteristics at baseline. Although also in our

study, multivariable analyses were performed, these multivariable

adjustments can not correct for all patient characteristics, and does not rule

out the effect of selection of patients who are candidates for

revascularisation, and those who are not.

Diabetes was an independent predictor of long-term mortality in our total

study population, but in particular in patients scheduled for medical

treatment or PCI. The better prognosis of patients with diabetes who were

treated by CABG may be due to more complete revascularisation [18],

particularly if the internal mammarian artery (IMA) is used [19]. PCI may be

less effective in diabetic patients because of smaller caliber vessels and a

high incidence of restenosis [20]. In contrast, the long-term patency of an

IMA after CABG is comparable between patients with and without diabetes

Chapter 5 ___________________________________________________________________

79

[21]. Furthermore, patients treated with PCI in our study, were treated only

with balloon, and especially in patients with diabetes use of (drug-eluting)

intracoronary stents may improve prognosis after PCI.

In all subgroups in our study, an impaired left ventricular function was

associated with increased long-term mortality. Apart from treatment of

myocardial ischemia by revascularisation, medical treatment may be of

benefit in these patients, including beta-blockers and ACE-inhibitors.

Probably, use of implantable cardiodefibrillators can also improve long-term

survival in these patients [22].

Peripheral vascular disease was a predictor of long-term mortality,

particularly in patients scheduled for PCI. In the total study group peripheral

vascular disease was not an independent significant predictor of long-term

mortality after multivariable analyses. Peripheral vascular disease may be

associated with more extensive and diffuse coronary artery disease [23].

Limitations

Our study has several limitations. Because the sample size of the study

was not very large, the number of patients in the subgroups were small. We

had no data on major cardiac events as myocardial infarction or additional

revascularisation during the follow-up period. There was also no

information on use of medication or anginal status during the follow-up

period. The cause of death could not be traced in our majority of patients.

Of course, during the 10-year follow-up period, treatment strategies

changed and new interventions were introduced. CABG and PCI

techniques improved, including off-pump surgery and the use of (coated)


80

stents and GP IIb/IIIa inhibitors in PCI. Medical treatment improved too,

with more frequent use of statins, beta-blockers and ACE-inhibitors. We did

an observational study, and some of our observations, including mortality

differences between invasive and conservative treated patients may have

been caused by patient selection.

Conclusions

1. In patients with coronary artery disease in who an invasive approach is

considered, increasing age, decreased left ventricular function and diabetes

were the most strong predictors of long-term mortality.

2. Invasive treatment was not associated with lower long-term mortality

after multivariable analyses. Although in the analyses adjustments were

made for several potential confounders, these findings may, however, be

caused by patient selection.

Chapter 5 ___________________________________________________________________

81

References

1. SAPAT (Swedish Angina Pectoris Aspirin Trial) Group, Juul-Moller S, Edvardsson N, Jahnmatz B, et al. Double-blind trial of aspirin in primary prevention of yocardial infarction in patients with stable chronic angina pectoris. Lancet 1992;340:1421-5. 2. Ridker PM, Manson JE, Gaziano JM, et al. Low-dose aspirin therapy for chronic stable angina: A randomized, placebo-controlled clinical trial. Ann Intern Med 1991;114:835-9. 3. Scandinavian Simvastatin Survival Study Goup. Randomised trial of cholesterol lowering in 4444 patients with coronary artery disease: the Scandinavian Simvastatin Survival Study (4S). Lancet 1994;344:1383-9. 4. Pedersen TR, Wilhelmsen L, Faergeman O, et al. Follow-up study of patients randomized in the Scandinavian simvastatin survival study (4S) of cholesterol lowering. Am J Cardiol 2000;86:257-62. 5. Chaitman BR, Ryan TJ, Kronmal RA, Foster ED, et al. Coronary Artery Surgery Study (CASS): comparability of 10 year survival in randomized and randomizable patients. J Am Coll Cardiol. 1990;16:1071-8. 6. The BARI Investigators. Seven-year outcome in the Bypass Angioplasty Revascularization Investigation (BARI) by treatment and diabetic status. J Am Coll Cardiol 2000;35:1122-9. 7. Rodriguez A, Mele E, Peyregne E, et al. Three-year follow-up of the Argentine Randomized Trial of Percutaneous Transluminal Coronary Angioplasty Versus Coronary Artery Bypass Surgery in Multivessel Disease (ERACI). J Am Coll Cardiol 1996;27:1178-84. 8. King SB III, Kosinski AS, Guyton RA, et al. Eight-year mortality in the Emory Angioplasty versus Surgery Trial (EAST). J Am Coll Cardiol 2000;35:1116-21. 9. Hoffman SN, TenBrook JA, Wolf MP, et al. A meta-analysis of randomized controlled trials comparing coronary artery bypass graft with percutaneous transluminal coronary angioplasty: one- to eight-year outcomes. J Am Coll Cardiol 2003;41:1293-304. 10. Emond M, Mock MB, Davis KB, et al. Long-term survival of medically treated patients in the Coronary Artery Surgery Study (CASS) Registry. Circulation 1994;90:2645-57. 11.Jones RH, Kesler K, Phillips HR 3rd, et al. Long-term survival benefits of coronary artery bypass grafting and percutaneous transluminal angioplasty in patients with coronary artery disease. J Thorac Cardiovasc Surg 1996;111:1013-25. 12.Rigter H, Meijler AP, McDonnell J, et al. Indications for coronary revascularisation: a Dutch perspective. Heart 1997;77:211-8.


82

13. Meijler AP, Rigter H, Bernstein SJ, et al. The appropriateness of intention to treat decisions for invasive therapy in coronary artery disease in The Netherlands. Heart 1997;77:219-24. 14. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457-80. 15. Cox DR. Regression models and life tables. J R Stat Society 1972:34;187- 220. 16. Henderson RA, Pocock SJ, Clayton TC, et al. Second Randomized Intervention Treatment of Angina (RITA-2) Trial Participants. Seven-year outcome in the RITA-2 trial: coronary angioplasty versus medical therapy. J Am Coll Cardiol. 2003;42:1161-70. 17. Yusuf S, Flather M, Pogue J, et al. Variations between countries in invasive cardiac procedures and outcome in patients with suspected unstable angina or myocardial infarction without initial ST elevation. Lancet 1998;352:507-14. 18. Zhao XQ, Brown BG, Stewart DK et al. Effectiveness of revascularization in the Emory angioplasty versus surgery trial. A randomized comparison of coronary angioplasty with bypass surgery. Circulation 1996;93:1954-62. 19. Schwartz L, Kip KE, Frye RL et al. Coronary bypass graft patency in patients with diabetes in the Bypass Angioplasty Revascularization Investigation (BARI). Circulation 2002;106:2652-58. 20. Van Belle E, Bauters C, Hubert E et al. Restenosis rates in diabetic patients: a comparison of coronary stenting and balloon angioplasty in native coronary vessels. Circulation 1997;96:1454-60. 21. Detre KM, Lombardero MS, Brooks MM et al. The effect of previous coronary- artery bypass surgery on the prognosis of patients with diabetes who have acute myocardial infarction. Bypass Angioplasty Revascularization Investigation Investigators. N Engl J Med 2000;342:989-97. 22. Moss AJ, Zareba W, Hall WJ, et al. Prophylactic implantation of a defibrillator in patients with myocardial infarction and reduced ejection fraction. N Engl J Med 2002;346:877-83. 23. Sukhija R, Yalamanchili K, Aronow WS. Prevalence of left main coronary artery disease, of three- or four-vessel coronary artery disease, and of obstructive coronary artery disease in patients with and without peripheral arterial disease undergoing coronary angiography for suspected coronary artery disease. Am J Cardiol. 2003;92:304-5.

CHAPTER 6

Long-term survival after coronary revascularisation in patients

with diabetes

J.R. Timmer, A. Breeman, J.P. Ottervanger

E. de Kluiver, P. Boonstra, F. Zijlstra

Submitted

Chapter 6 ___________________________________________________________________

85

Abstract

Aims: The optimal method of revascularisation in diabetic patients with

coronary artery disease remains controversial. It was our aim to evaluate

long-term outcome of diabetic patients with coronary artery disease in daily

practice in whom an invasive approach was considered.

Methods: It concerns a prospective follow-up study of patients with

coronary artery disease in whom a coronary revascularisation procedure

was considered. Follow-up data were obtained on the vital status up to ten

years after inclusion.

Results: Of the 872 included patients, a total of 107 patients (12%) had

diabetes. Patients with diabetes were older and more frequently female.

Long-term mortality was higher in diabetes compared to non-diabetes (36%

vs. 25%, p=0.01). This association was observed in both medical treated

patients (65% vs. 31%, p=0.01) and in those treated with percutaneous

coronary intervention (41% vs. 24%, p=0.02). There was, however, no

difference in mortality in diabetes versus non-diabetes after coronary artery

bypass grafting (24% vs. 24%, p=0.89). Multivariate analysis did not

change these findings.

Conclusions: In patients with coronary artery disease in whom an invasive

approach is considered, diabetes is associated with a higher long-term

mortality. After coronary artery bypass grafting, long-term survival may be

comparable between patients with and without diabetes.

Long-term survival, importance of diabetes ___________________________________________________________________

86

Introduction

Diabetes is associated with impaired outcome after coronary

revascularisation [1,2] The optimal method of revascularisation remains

controversial. In the BARI trial, coronary artery bypass grafting (CABG)

when compared to percutaneous coronary intervention (PCI), was

associated with a significant 7-year mortality reduction in patients with

diabetes [3]. However, this long-term benefit was not confirmed by other

randomised trials [4,5] Furthermore, patients in daily practice may differ

from those included in randomised trials, as was shown in the registry of

the BARI trial [6]. To evaluate long-term outcome of diabetic patients with

coronary artery disease (CAD) in daily practice in whom an invasive

approach was considered, we did a follow-up study of patients included in

the DUCAT (DUtch inventory of invasive Coronary Atherosclerosis

Treatments) study.

Patients and Methods

The DUCAT study (DUtch inventory of invasive Coronary Atherosclerosis

Treatments) was initiated in 1992 with the purpose to determine how

appropriate treatment decisions are concerning invasive treatment of

patients with CAD [7]. Assessment of appropriateness of medical decisions

was achieved using the RAND/UCLA method [8]. Six cardio-pulmonal

surgeons and 6 interventional cardiologists of 12 heart centres in the

Netherlands were asked to participate in a panel to determine

appropriateness of treatment decisions of all consecutive patients

Chapter 6 ___________________________________________________________________

87

presented to 10 heart teams consisting of at least one surgeon and one

interventional cardiologist. This panel method has been proven to be

consistent and reliable in assessing appropriateness [8].

Enrolment began in February 1992 for a period of 3 months. Each case

was presented by a clinical cardiologist in person or by letter, fax, or

telephone, and eventually led to an intention to treat decision in favour of

PCI, CABG, or medical treatment. All presentations were based on clinical

data and coronary angiographic results. Enrolment was approved for all

patients with significant CAD, defined by the DUCAT panel as a minimum

of 50% narrowing of the left main artery in left main disease, at least one

artery with 70% narrowing and other arteries with 50% narrowing in

multivessel disease, and one artery with 70% narrowing in one-vessel

disease. Patients who previously had CABG or in whom CABG was to be

combined with other surgery (cardiac or general) were excluded. During

presentation at the heart team, several variables were collected, including

demographics, medical history, risk factors for CAD, symptomatology,

medication, ischemia detection tests, coronary angiographic evaluation, left

ventricular function tests, urgency status and intention of treatment. The

main findings of the DUCAT study were published previously [7,8]. In

summary, 3646 consecutive patients were included. Unstable angina was

the most appropriate clinical status for intervention, whereas asymptomatic

coronary disease was the least.

The present study is a two centre follow-up study of all patients

consecutively presented to the cardio-pulmonal team of the heart centres of

Zwolle and Groningen in the Netherlands. In these two centres, a total of

1047 patients were presented to the cardio-pulmonal teams during the

study period. Follow-up data were obtained on the vital status up to ten


88

years after inclusion. Follow-up data were collected via the register office,

the General Practitioner or via a direct contact with the patient or his

relatives by telephone.

Statistics

Differences between group means were tested by two-tailed Student's

t-test. A Chi-square statistic was calculated to test differences between

proportions. Survival functions were calculated, using the Kaplan-Meier

product limit method. Mantel-Cox (or log-rank) test was applied to evaluate

the differences between survival functions.


assess the independent relation between revascularisation strategy and 10-

year survival after adjustment for baseline characteristics.

Results

Patients

Of the 1057 included patients, 10 cases were censored because they were

presented to the cardio-pulmonal team for the second time. Of the

remaining 1047 patients, 877 (84%) met the inclusion criteria and were

enrolled in the study. Of 5 patients the diabetic status was unknown, these

patients were not included in our analysis. Our analysis consisted therefore

of 872 patients. A total of 107 patients (12%) had diabetes. Patients with

diabetes were older (66 8 vs. 62 10 year, p<0.001), were more frequently

female (55% vs. 21%, p<0.001) and had a higher prevalence of

hypertension (46% vs. 27%, p<0.001) when compared to patients without

Chapter 6 ___________________________________________________________________

89

diabetes. Furthermore, patients with diabetes had more often multi vessel

disease. Differences in baseline characteristics between patients with and

without diabetes are shown in table 1.

Table 1: Baseline characteristics of all patients according to diabetic status.

Diabetes Non-diabetes p value N: 107 765

Age in yr. (mean (SD)) 66 8 62 10 < 0.001 Male 48 (44.9%) 601 (78.6%) < 0.001 Risk factors Hypertension 49 (45.8%) 208 (27.3%) < 0.001 Smoking 17 (18.9%) 209 (29.7%) 0.033 Lipid disorder 35 (35.7%) 261 (37.1%) 0.786 Family history 24 (28.9%) 278 (42.4%) 0.018 Obesity 17 (21.3%) 77 (13.2%) 0.054 Coronary history Previous MI 35 (32.7%) 292 (38.3%) 0.266 Previous PCI 13 (12.1%) 83 (10.9%) 0.694 Co-morbidity COPD 12 (11.2%) 66 (8.7%) 0.389 Periph Vasc Disease 17 (15.9%) 70 (9.2%) 0.030 Stroke 12 (11.2%) 41 (5.4%) 0.018 Vessels diseased One vessel 28 (26.2%) 275 (35.9%) 0.047 Two vessel 36 (33.6%) 214 (28.0%) 0.224 Three vessel 37 (34.6%) 223 (29.2%) 0.250 Left Main 6 (5.6%) 53 (6.9%) 0.610 Type C lesion 58 (54.2%) 382 (49.9%) 0.408 LVEF 20 – 40% 17 (16.3%) 116 (15.6%) 0.843 <20% 5 (4.8%) 20 (2.7%) 0.218

COPD = chronic obstructive pulmonary disease, MI = myocardial infarction, PCI = percutaneous coronary intervention, SD = standard deviation.


90

Coronary revascularisation

In patients without diabetes, PCI was recommended in 333 patients (44%),

CABG in 333 patients (44%) and a conservative approach was advised in

99 patients (13%). In patients with diabetes the type of revascularisation

was comparable to those without diabetes, with percentages of patients

recommended for PCI, CABG or conservative approach of 36%, 48% and

16%, respectively. There were several differences between diabetic

patients treated conservatively and treated invasively. Patients with

diabetes treated with PCI had more often single vessel disease (56% vs.

4%, p<0.001) compared to those treated with CABG. LVEF was

comparable between PCI and CABG treated patients.

Long-term mortality

In the total study group, long-term mortality was associated with increasing

age, revascularisation, diabetes (36% vs. 25%, p=0.01), peripheral artery

disease and a LVEF < 20%. As compared to revascularisation by either

PCI or CABG, medical treated patients had an increased mortality. The

increased mortality in patients with diabetes was observed in both medical

treated patients (n=116) (diabetes 65% vs. non-diabetes 31%, p=0.01) and

in those who received PCI (n=372) (diabetes 41% vs. non-diabetes 24%,

p=0.02). There was, however, no difference in mortality between diabetes

and non-diabetes after CABG (n=384) (24% vs. 24%, p=0.89). Survival

curves of patients with and without diabetes according to type of

revascularisation are shown in Figures 1-3.

Chapter 6 ___________________________________________________________________

91

Multivariate analysis

To study the independent prognostic importance of diabetes on clinical

outcome, multivariate analysis was performed. We included age, gender

and all univariate predictors of long term mortality in the multivariate model.

After multivariate analysis increasing age, diabetes and a decreased LV

Figure 1: 10-year follow-up of 116 medically treated patients.

Figure 2: 10-year follow-up of 372 PCI treated patients.

Figure 3: 10-year follow-up of 384 CABG treated patients.

1 2

3


92

function were independent predictors of long-term mortality (table 2). The

independent predictive value of diabetes was most pronounced in patients

treated without revascularisation and in those treated with PCI (OR 2.1;

95% CI 1.3 to 3.3). In patients treated with CABG, diabetes was no

independent predictor of long term mortality (OR 1.2; 95% CI 0.6 to 2.4).

Discussion

In patients with coronary artery disease in whom an invasive approach is

considered, diabetes is associated with increased long term mortality.

However, in patients treated with CABG this association was not observed.

Our analysis reflects real world clinical practice and has additional value

regarding the still ongoing debate about the optimal method of

revascularisation of patients with diabetes.

Table 2: Predictors of mortality in patients in the DUCAT study.

OR 95% CI Male gender 1.36 0.98 to 1.89 Age (per year) 1.09 1.07 to 1.12 Conservative treatment# 1.24 0.86 to 1.79 Diabetes 1.46 1.01 to 2.13 Periph. Vasc. Disease 1.28 0.85 to 1.91 Decreased LV function 1.58 1.15 to 2.18

#As compared to revascularisation by either PCI or CABG. Adjusted for differences in the other variables. DUCAT = DUtch inventory of invasive Coronary Atherosclerosis Treatments.

Chapter 6 ___________________________________________________________________

93

Diabetes versus non-diabetes

Patients with diabetes had a higher long term mortality compared to

patients without diabetes. This could be due to differences in baseline

characteristics. In general, patients with diabetes were older, were more

often female and had a higher prevalence of hypertension. However, after

multivariate analysis diabetes was still a significant predictor of mortality.

There may be several mechanisms for this increase in mortality. A

procoaguable state and more unfavourable lipid levels in diabetic patients

might play a role [9,10]. Atherosclerotic coronary abnormalities may be

more progressive in diabetes. Furthermore, pre existing left ventricular

failure, either diastolic or systolic, may contribute to an increase in mortality

[11]. Moreover, glycometabolic disturbances during acute coronary events

increase infarct size and might predispose to ventricular arrhythmia’s [12].

Interestingly, in patients with extensive CAD, whom are generally treated

with CABG, the difference in mortality between diabetes and non-diabetes

seems to be much less clear [13].

PCI as revascularisation strategy

Mortality was higher in PCI treated patients with diabetes compared to

those without diabetes. There may be several explanations for this

difference. Patients with diabetes have smaller calibre vessels and higher

rates of restenosis than patients without diabetes [14]. Furthermore, more

progressive atherosclerosis in diabetes may also affect coronary segments

not significantly stenosed at the time of the initial decision of

revascularisation. As diabetic patients already have a comprised LV

function, display impaired preconditioning and have glucometabolic

disturbances potentially increasing ischemic myocardial damage, they


94

might be more prone to die when suffering subsequent coronary events

[11,15].

CABG as revascularisation strategy

There was no difference in mortality in diabetic patients treated with CABG

compared to non-diabetic patients treated with CABG. Previous studies

show conflicting results. Several studies report that diabetes is associated

with a worse outcome after CABG [1,16]. The randomised EAST trial and

the recently published study of Calafiore et al. however, did not find an

association between diabetes and an long term adverse prognosis after

CABG [5,17].

Revascularisation in diabetes

In patients with diabetes, those treated medically had highest mortality,

whereas the lowest mortality was found in diabetic patients treated with

CABG. Differences in baseline characteristics between medical treated

patients and those treated with revascularisation, may partly explain the

differences in outcome. Medically treated patients were older and had a

higher prevalence of peripheral artery disease and COPD than those

treated with coronary revascularisation. The improved prognosis after

CABG in diabetic patients can be caused by the fact that PCI in diabetes is

associated with a higher restenosis rate, whereas graft patency after 4

years between patients with and without diabetes may be comparable.[18]

Also, incomplete revascularisation might be detrimental in diabetic patients

treated with PCI. Coronary artery bypass grafting may be superior in

reaching complete revascularisation leaving PCI treated patients with an

increased area at risk for future ischemic events [19]. Furthermore, there

Chapter 6 ___________________________________________________________________

95

seems to be a protective effect from the use of an internal thoracic artery

(ITA) in CABG, especially in patients with diabetes [20]. The mortality risk

after myocardial infarction in diabetic patients without ITA is much higher

when compared to diabetic patients who had received revascularisation

with an ITA [20]. In our study the majority of patients were treated with an

ITA.

Other studies

In the BARI trial, 1829 patients with multivessel disease were randomised

towards CABG or PCI [3]. This study found a better 7-year survival in

diabetic patients (n=353) treated with CABG compared to PCI (76% vs.

56%). However, other randomised trials, as the CABRI trial and the EAST

trial, did not find a significant beneficial effect after CABG compared to PCI

in diabetic patients [4,5]. Moreover, the RITA trial including 62 patients with

diabetes found a non-significant worse outcome for diabetic patients

randomised to CABG compared to PCI [21]. Observational studies also

found contradicting evidence regarding optimal method of reperfusion. The

Duke and EMORY analyses did not find a benefit of CABG versus PCI in

diabetic patients, although insulin treated diabetes seemed to benefit from

CABG [1,2]. The MAHI study found an unadjusted survival benefit of

diabetic patients when treated with CABG compared to PCI.[22]

Interestingly, in diabetic patients from the registry of the BARI trial there

was no significant difference in mortality between the two revascularisation

methods [6]. Niles et al. however did find a significant reduction in mortality

in more than 7000 diabetic patients treated with CABG versus PCI after 5

years follow-up [23]. Our study also did find a survival benefit of diabetic


96

patients when treated with CABG compared to PCI after a long follow-up

period of 10 years.

Study limitations

This was an observational study without randomisation. This could have led

to differences in unmeasured baseline characteristics for which no

correction or adjustment could be made.

Unfortunately, no information about the type of anti-diabetic treatment was

available. During the study period, intracoronary stenting and treatment

with glycoprotein IIb/IIIa inhibitors or clopidogrel were not available. These

new therapeutic modalities may well have a profound effect on clinical

outcome and may, in particular, improve clinical outcome in patients with

diabetes when treated with PCI [24,25].

Conclusion

Diabetic patients with significant CAD have a higher long-term mortality

compared to patients without diabetes. After CABG however, survival is

comparable between diabetic and non-diabetic patients. Complete

revascularisation may decrease the influence of diabetes on survival.

Chapter 6 ___________________________________________________________________

97

References

1. Barsness GW, Peterson ED, Ohman EM, Nelson CL, DeLong ER, Reves JG, et al. Relationship between diabetes and long-term survival after coronary bypass and angioplasty. Circulation 1997;96:2551-2556. 2. Weintraub WS, Stein B, Kosinski A, Douglas JS Jr, Ghazzal ZM, Jones EL et al. Outcome of coronary bypass surgery versus coronary angioplasty in diabetic patients with multivessel coronary artery disease. J Am Coll Cardiol 1998;31:10-19. 3. The BARI Investigators. Seven-year outcome in the Bypass Angioplasty Revascularization Investigation (BARI) by treatment and diabetic status. J Am Coll Cardiol 2000;35:1122-1129. 4. Kurbaan AS, Bowker TJ, Ilsley CD, Sigwart U, Rickards AF; On behalf of the CABRI Investigators. Difference in the mortality of the CABRI diabetic and nondiabetic populations and its relation to coronary artery disease and the revascularization mode. (Coronary Angioplasty versus Bypass Revasculari- zation Investigation). Am J Cardiol 2001;87:947-950. 5. King SB III, Kosinski AS, Guyton RA, Lembo NJ, Weintraub WS. Eight-year mortality in the Emory Angioplasty versus Surgery Trial (EAST). J Am Coll Cardiol 2000;35:1116-1121. 6. Detre KM, Guo P, Holubkov R, Califf RM, Sopko G, Bach R, Brooks et al. Coronary revascularization in diabetic patients: a comparison of the randomized and observational components of the Bypass Angioplasty Revascularization Investigation (BARI). Circulation 1999;99:633-640. 7. Rigter H, Meijler AP, McDonnell J, Scholma JK, Bernstein SJ. Indications for coronary revascularisation: a Dutch perspective. Heart 1997;77:211-218. 8. Meijler AP, Rigter H, Bernstein SJ, Scholma JK, McDonnell J, Breeman A et al. The appropriateness of intention to treat decisions for invasive therapy in coronary artery disease in The Netherlands. Heart 1997;77:219-224. 9. Winocour PD. Platelet abnormalities in diabetes. Diabetes 1992;41(Suppl 2): 26-31. 10. Kreisberg RA. Diabetic dyslipidemia. Am J Cardiol 1998;82:67U-73U. 11. Norhammar A, Malmberg K. Heart failure and glucose abnormalities: an increasing combination with poor functional capacity and outcome. Eur Heart J 2000;21:1293-1294. 12. Oliver MF, Opie LH. Effects of glucose and fatty acids on myocardial ischaemia and arrhythmias. Lancet 1994;343:155-158. 13. Ishihara M, Sato H, Kawagoe T, Shimatani Y, Kurisu S, Nishioka K et al. Impact of diabetes mellitus on long term survival after acute myocardial infarction in patients with single vessel disease. Heart 2001;86:133-138. 14. van Belle E, Bauters C, Hubert E, Bodart JC, Abolmaali K, Meurice T et al. Restenosis rates in diabetic patients: a comparison of coronary stenting and balloon angioplasty in native coronary vessels. Circulation 1997;96:1454- 1460.


98

15. Ishihara M, Inoue I, Kawagoe T, Shimatani Y, Kurisu S, Nishioka K et al. Diabetes prevents ischemic preconditioning in patients with a first acute anterior wall myocardial infarction. J Am Coll Cardiol 2001;38:1007-1011. 16. Brooks MM, Jones RH, Bach RG, Chaitman BR, Kern MJ, Orszulak TA et al. Predictors of mortality and mortality from cardiac causes in the bypass angioplasty revascularization investigation (BARI) randomized trial and registry. For the BARI Investigators. Circulation 2000;101:2682-2689. 17. Calafiore AM, Di Mauro M, Di Giammarco G, Contini M, Vitolla G, Iaco AL et al. Effect of diabetes on early and late survival after isolated first coronary bypass surgery in multivessel disease. J Thorac Cardiovasc Surg 2003;125: 144-154. 18. Schwartz L, Kip KE, Frye RL, Alderman EL, Schaff HV, Detre KM. Coronary bypass graft patency in patients with diabetes in the Bypass Angioplasty Revascularization Investigation (BARI). Circulation 2002;106:2652-2658. 19. Zhao XQ, Brown BG, Stewart DK, Hillger LA, Barnhart HX, Kosinski AS et al. Effectiveness of revascularization in the Emory angioplasty versus surgery trial. A randomized comparison of coronary angioplasty with bypass surgery. Circulation 1996;93:1954-1962. 20. Detre KM, Lombardero MS, Brooks MM, Hardison RM, Holubkov R, Sopko G et al. The effect of previous coronary-artery bypass surgery on the prognosis of patients with diabetes who have acute myocardial infarction. Bypass Angioplasty Revascularization Investigation Investigators. N Engl J Med 2000; 342:989-997. 21. Henderson RA, Pocock SJ, Sharp SJ, Nanchahal K, Sculpher MJ, Buxton MJ et al. Long-term results of RITA-1 trial: clinical and cost comparisons of coronary angioplasty and coronary-artery bypass grafting. Randomised Intervention Treatment of Angina. Lancet 1998;352:1419-1425. 22. Gum PA, O'Keefe JH, Jr., Borkon AM, Spertus JA, Bateman TM, McGraw JP et al. Bypass surgery versus coronary angioplasty for revascularization of treated diabetic patients. Circulation 1997;(Suppl II);96:7-10. 23. Niles NW, McGrath PD, Malenka D, Quinton H, Wennberg D, Shubrooks SJ et al. 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;37:1008-1015. 24. van Belle E, Perie M, Braune D, Chmait A, Meurice T, Abolmaali K et al. Effects of coronary stenting on vessel patency and long-term clinical outcome after percutaneous coronary revascularization in diabetic patients. J Am Coll Cardiol 2002;40:410-417. 25. Lincoff AM, Tcheng JE, Califf RM, Kereiakes DJ, Kelly TA, Timmis GC et al. Sustained suppression of ischemic complications of coronary intervention by platelet GP IIb/IIIa blockade with abciximab: one-year outcome in the EPILOG trial. Evaluation in PCI to improve long-term outcome with abciximab GP IIb/IIIa blockade. Circulation 1999;99:1951-1958.

CHAPTER 7

Treatment decisions in patients with

stable coronary artery disease in a broad

range of European practices.

Insights from the Euro Heart Survey on

coronary revascularization

A. Breeman, M. Hordijk, M. Lenzen, S. Hoeks, J.P. Ottervanger,

M. Bertrand, U. Sechtem, R. Zaliunas, V. Legrand, M.J. de Boer,

E. Stahle, N. Mercado, W. Wijns, E. Boersma

J Thor Cardiov Surg (in press)

Chapter 7 ___________________________________________________________________

101

Abstract

Objectives: To assess determinants of clinical decision-making in patients

with stable coronary artery disease.

Methods: The 2936 patients with stable angina pectoris who enrolled the

Euro Heart Survey on Coronary Revascularization were subject of this

analysis. After the diagnosis has been confirmed, physicians decided upon

treatment: medical management, or revascularization therapy by

percutaneous coronary intervention (PCI) or coronary bypass surgery

(CABG). We applied logistic regression analyses to evaluate the relation

between baseline characteristics and treatment decision: medical treatment

versus PCI, medical treatment versus CABG and PCI versus CABG.

Results: Median age was 64 years, 77% were men and 20% had diabetes.

Medical therapy was intended in 690 (24%), PCI in 1503 (51%), and CABG

in the remaining 743 (25%) patients respectively. Revascularization was

generally preferred in patients with more severe anginal complaints, an

intermediate to large area of myocardium at risk, and a preserved left

ventricular function, who had not undergone prior coronary

revascularization, provided lesions were suitable for treatment. CABG was

preferred over PCI in multivessel or left main disease, as well as in those

with concomitant valvular heart disease, provided a sufficient number of

lesions were suitable for CABG. In those with previous CABG, more often

PCI was prefered than re-CABG. Diabetes was not associated with more

often preference for CABG.

Conclusion: The EHS-CR revealed that treatment decisions in stable CAD

were largely in agreement with professional guidelines and determined by

multiple factors. Most important deviations between guideline

Treatment decisions in stable angina ___________________________________________________________________

102

recommendations and clinical practice were seen in patients with extensive

coronary disease, impaired left ventricular function and those with diabetes.

Introduction

In the Western World, stable coronary disease is a common and disabling

disorder. Treatment aims to minimize or abolish symptoms, to prevent

disease progression, and to improve prognosis by reducing the risk of

myocardial infarction or death. According to current treatment guidelines

patients should receive cardio-protective pharmaco-therapy, including anti-

platelet, anti-thrombotic and lipid-lowering agents, as well as anti-anginal

medication. In selected patients, coronary revascularization by either

percutaneous coronary intervention (PCI) or coronary artery bypass

grafting (CABG) may be appropriate. In the absence of symptomatic

indication, revascularization is recommended in patients with left main

disease, multivessel disease, significant disease of the proximal part of the

left anterior descending (LAD) artery, and in patients with a fair amount of

viable myocardium at risk. CABG is preferred in patients with diabetes,

those with more extensive disease, impaired left ventricular function, and in

patients with anatomy that is not suitable for PCI [1-11].

The European Society of Cardiology (ESC) realized that the clinical

management of cardiac patients in daily practice may vary from treatment

guidelines. A broad range of clinical and lesion-specific characteristics may

play a role in shaping treatment decisions in the real world. In order to

obtain quantitative information on the adherence to European and

international guidelines across member states the ESC has launched the

Chapter 7 ___________________________________________________________________

103

Euro Heart Survey (EHS) program (extensive information regarding this

program is available on the www) [12]. The EHS on coronary

revascularization (EHS-CR), which was conducted in 2001-2002, with

follow-up in 2003, enrolled 3006 consecutive patients with documented

stable coronary disease undergoing coronary angiography. This well-

characterized study population provides a unique opportunity to conduct a

systematic analysis of the relation between patient characteristics and

choices about patient management.

Methods

Patients

The EHS-CR has been described in detail elsewhere [12,13]. Briefly, the

survey was designed to screen all consecutive patients undergoing

invasive coronary diagnostic or therapeutic procedures. Patients were

enrolled if they had a diameter stenosis of at least 50% in at least one

major epicardial coronary artery. Data were then sent to a central database

in the European Heart House (Sophia Antipolis, France) via the www, using

the MacroTM software (InferMed, UK). The collected data included

demographics, co-morbidity, diagnosis, and detailed information regarding

diagnostic angiography and treatment modalities. Clinical outcome and

complications were recorded during the initial hospitalization and at one-

year follow-up. Via the applied software, data were automatically checked

for completeness, internal consistency and accuracy. The data

management staff at the European Heart House performed additional edit

checks. If necessary, queries were resolved with the local data collecting


104

officers. Between November 1, 2001 and March 1, 2002 a total of 5767

patients were enrolled. Follow-up was obtained at 11 months after

enrolment. The current analysis is limited to the 3006 (52%) patients with

stable angina as primary diagnosis.

Treatment decisions

The EHS-CR is a descriptive study, and the survey protocol did not dictate

any treatment decision. In general, physicians were encouraged to treat

their patients in conformance with the most recent guidelines. In order to be

informed of the physicians preferred, intended treatment, the survey

included the following question: As the treating physician, which treatment

option would be your first choice? According to the reply to this question,

patients were classified with a physician’s intention for medical treatment,

PCI, or CABG.

Data description and data analysis

Continuous data are described as median values with corresponding 25th

and 75th percentiles, and dichotomous data are described as counts and

percentages. Differences in baseline clinical and angiographical

characteristics in subgroups of patients according to intended treatment

were analyzed by unpaired Student’s t-tests, analyses of variance

(continuous data) and Chi-square tests (dichotomous data) as appropriate.

All statistical tests were 2-sided, and significance was stated at P=0.05.

Altogether 893 patients (30%) had missing data on at least one of the

variables that were considered as potential determinants of treatment

decisions (table 1). Simple missing data imputation was therefore applied,

by assigning the series mean value of the variable at hand. Univariable and

Chapter 7 ___________________________________________________________________

105

multivariable logistic regression analyses were then applied to further

evaluate the relation between a broad range of baseline characteristics and

intended treatment. Separate regression models were developed for

intended medical treatment versus PCI, medical treatment versus CABG,

and CABG versus PCI. All variables entered the multivariable stage,

irrespective of the results of univariable analyses. The final multivariable

model was then constructed by backward deletion of the least significant

characteristics, while applying P=0.05 as the threshold of significance.

The performance of the multivariable models was studied with respect to

discrimination and calibration. Discrimination refers to the ability to

distinguish patient subgoups (intended medical treatment versus PCI;

intended medical treatment versus CABG; intended CABG versus PCI) by

using the model. It was quantified by a measure of concordance, the c-

statistic. For binary outcomes the c-statistic is identical to the area under

the receiver operating characteristic curve. The c-statistic lies between 0.5

and 1, and is better if closer to 1 [14]. Calibration refers to whether the

predicted frequencies (by using the model) of patients with a phycisian’s

preference for a specific treatment modality agree with the observed

frequencies. Calibration was measured with the Hosmer-Lemeshow

goodness-of-fit test [15,16].


106

Table 1: Clinical characteristics of the study population.

All Intended treatment P Medical PCI CABG No. of patients 2936 690 1503 743

Demographics Age, years 64 (56-71) 64 (56-71) 63 (54-70) 66 (57-72) ‡ Men 77 75 76 80 §

Risk factors, medical history and medication at admission Current smoking 21 17 23 21 § Diabetes mellitus (patients using insulin or oral antidiabetic drugs) 20 22 19 21

Hypercholesterolemia 69 67 71 66 § Hypertension 61 64 60 61 PCI 24 27 28 12 ‡ CABG 13 24 12 4 ‡ Myocardial infarction 43 49 41 40 † Peripheral vascular disease 13 13 11 17 ‡ Cerebro-vascular disease 8 8 6 10 § Renal insufficiency 3 5 4 2 § Use of beta-blockers, calcium antagonists, or nitrates †

None 11 13 10 11 Mono 36 40 37 32 Double 41 37 42 45 Triple 11 10 12 12

Presentation Concomitant valvular heart disease 7 8 4 13 ‡ NYHA symptomatic (heart failure) class ‡ No heart failure 84 80 89 80 I or II 10 13 8 13 III or IV 5 8 3 7 CCS angina class ‡ I 15 24 14 10 II 48 49 50 42 III 32 25 31 41 IV 5 2 5 6 Size of the myocardium at risk * ‡ Small 28 39 29 15 Intermediate 53 44 54 58 Large 19 16 17 27 EuroSCORE ¶ 3 (2-5) 4 (2-6) 3 (1-5) 4 (2-6) ‡

Chapter 7 ___________________________________________________________________

107

Table 1: continued All Intended treatment P Medical PCI CABG Left ventricular function # ‡ Normal (LVEF >50%) 66 59 71 62 Mild impairment (LVEF 40-50%) 23 23 22 26 Moderate impairment (LVEF 30-40%) 7 10 5 9 Severe impairment (LVEF <30%) 4 8 2 3

Continuous data (age, EuroSCORE) are presented as median values (25th - 75th percentile); dichotomous data are presented as percentages. P-values: § <0.05; † <0.01; ‡ <0.001. * Qualitative estimate based on non-invasive diagnostics as described in the ACC/AHA

guidelines [10]. ¶ European System for Cardiac Operative Risk Evaluation, which is a score developed to

quantify the risk of perioperative mortality in patients scheduled for cardiac surgery [22]. # Based on quantitative or qualitative measurements.

Results

Patients and intended treatment

The intended treatment was not recorded in 70 (2.3%) patients, leaving

2936 patients suitable for analysis. Medical therapy was intended in 690

(24%), PCI in 1503 (51%), and CABG in the remaining 743 (25%) patients

respectively. Eighty-six percent of patients who were selected for coronary

revascularization underwent this procedure within one year after initial

coronary angiography. Twelve percent of patients undergoing PCI were

treated for in-stent restenosis or restenosis after a prior non-stent

procedure. In 15% of patients undergoing CABG, coronary surgery was

combined with valve repair or replacement.


108

Table 2: Angiographical characteristics of the study population.

All Intended treatment P Medical PCI CABG

No. of patients 2936 690 1503 743 Mitral insufficiency potentially requiring surgery

14 16 11 17 ‡

Diseased arteries

Left Anterior Descending 73 69 67 88 ‡ Proximal Left Anterior Descending 33 26 26 52 ‡ Left Circumflex 59 58 51 78 ‡ Right Coronary Artery 64 64 56 80 ‡ Left main 9 8 4 21 ‡

Number of diseased arteries ‡ 1 34 40 44 8 2 30 26 35 21 3 27 25 17 50 LM 9 8 4 21 Number of diseased segments ‡ 1 29 34 38 8 2 23 21 28 15 3 18 17 16 21

4 30 28 18 57 Total occlusions ‡ 0 59 46 69 51 1 26 31 21 31 2 9 13 6 12

3 6 11 4 6 Type C lesions ‡ 0 56 48 63 48 1 23 24 23 22 2 11 14 8 15

3 11 14 7 14 Lesions suitable for PCI ‡ 0 28 56 6 46 1 37 22 53 16 2 20 12 26 15

3 16 10 15 23

Chapter 7 ___________________________________________________________________

109

Table 2: continued All Intended treatment P Medical PCI CABG Lesions suitable for CABG ‡ 0 24 41 26 4 1 26 21 37 8 2 17 13 20 16

3 33 24 17 72 Grafts with >50% diameter stenosis * ‡ 0 39 50 30 27 1 30 28 34 27

2 31 22 36 46

Data represent percentages. P-values: § <0.05; † <0.01; ‡ <0.001. * In patients with a history of prior CABG only.

Clinical and angiographical characteristics

The median age of the study population was 64 years, and 77% were men.

Twenty percent of patients had diabetes. A history of prior PCI was

reported in 24% of patients and a prior CABG in 13%. Most patients had

mild to moderate anginal complaints and an intermediate risk profile. A

moderately or severely impaired left ventricular function was observed in

11%. Thirty-four percent of patients had single vessel disease, 57% had

multivessel disease, and 9% had left main disease.

Important differences in relevant clinical (table 1) and angiographic (table 2)

baseline characteristics were observed according to intended treatment.

Patients in whom the physician preferred medical therapy more often had a

history of coronary revascularization or a history of myocardial infarction

than the remaining patients. Their anginal complaints were relatively mild,

but they more often had symptomatic heart failure. Patients scheduled for

medical therapy had a small area of jeopardized myocardium, a high

number of total occlusions and type C lesions, and a low number of lesions

suitable for revascularization.


110

Table 3: Association between patient characteristics and the phycisian’s preference for PCI versus medical treatment, CABG versus medical treatment, and CABG versus PCI.

Intended treatment PCI versus medical CABG versus medical CABG versus PCI

Odds ratio (95% CI) §

2 Odds ratio (95% CI) †

2 Odds ratio (95% CI) ‡

2

Age, year <60 1 60-70 0.62 (0.41, 0.94) 5

70 0.42 (0.25, 0.69) 12 Male gender 1.4 (1.1, 1.8) 5 1.6 (1.1, 2.3) 6

Hypercholesterolemia 0.62 (0.45, 0.85) 9

Prior PCI 0.65 (0.43, 0.99) 4 0.63 (0.43, 0.93) 5 Prior CABG No 1 1 1 Yes, no diseased grafts 0.43 (0.27, 0.70) 11 0.02 (0.01, 0.05) 66 0.07 (0.02, 0.22) 21 Yes, 1 diseased graft 1.5 (1.0, 2.4) 3 0.04 (0.02, 0.07) 86 0.02 (0.01, 0.04) 88 Prior myocardial infarction 0.75 (0.54, 1.0) 3

Renal insufficiency 0.28 (0.12, 0.66) 9 0.21(0.08, 0.53) 11 Concomitant valvular heart disease

4.8 (2.7, 8.4) 30 5.9 (3.7, 9.9) 44

NYHA symptomatic (heart failure) class

No heart failure 1 I or II 1.0 (0.62, 1.6) 0 III or IV 0.46 (0.25, 0.84) 6

CCS angina class I 1 1 II 1.8 (1.3, 2.5) 13 1.6 (1.0, 2.6) 4 III 1.8 (1.3, 2.6) 11 2.4 (1.4, 3.9) 11 IV 6.2 (2.7, 14) 19 7.1 (2.5, 20) 13

Non-invasive risk stratification

Low 1 1 Intermediate 2.0 (1.6, 2.7) 27 3.3 (2.2, 4.9) 35 High 2.4 (1.6, 3.6) 18 3.4 (2.0, 5.7) 20 EuroSCORE 3 0.71 (0.54, 0.93) 6 2.1 (1.3, 3.4) 10 2.0 (1.5, 2.7) 19

Chapter 7 ___________________________________________________________________

111

Table 3: continued Intended treatment PCI versus medical CABG versus medical CABG versus PCI

Odds ratio (95% CI) §

2 Odds ratio (95% CI) †

2 Odds ratio (95% CI) ‡

2

Left ventricular function Normal (LVEF >50%) 1 1 Mild impairment (LVEF 40-50%)

1.0 (0.79, 1.3) 0 0.75 (0.52, 1.1) 3

Moderate impairment (LVEF 30-40%)

0.38 (0.24, 0.61) 16 0.49 (0.27, 0.87) 6

Severe impairment (LVEF <30%)

0.27 (0.14, 0.52) 15 0.18 (0.08, 0.42) 16

Proximal left anteriordescending disease

2.2 (1.6, 3.1) 22 1.9 (1.4, 2.6) 17

Left main disease 6.6 (3.4, 12) 34 12 (6.3, 22) 60 Number of diseased arteries 1 1 1 2 2.1 (1.2, 3.5) 8 1.4 (0.85, 2.2) 2 3 3.9 (2.2, 6.8) 23 5.4 (3.2, 8.9) 43 Total occlusions 0 1 1 1 0.69 (0.52, 0.91) 7 1.3 (0.93, 1.8) 2 2 0.56 (0.36, 0.87) 7 1.1 (0.67, 1.9) 0 3 1.0 (0.60, 1.8) 0 0.42 (0.21, 0.87) 6 Type C lesions 0 1 1 0.87 (0.59, 1.3) 0 2 0.73 (0.45, 1.2) 2 3 0.40 (0.24, 0.66) 12 Lesions suitable for PCI 0 1 1 1 1 28 (20, 41) 308 0.55 (0.36, 0.85) 7 0.03 (0.02, 0.05) 150 2 35 (22, 56) 232 0.46 (0.28, 0.75) 10 0.01 (0.01, 0.02) 185

3 41 (24, 69) 189 0.40 (0.24, 0.66) 13 0.01 (0.01, 0.02) 185 Lesions suitable for CABG 0 1 1 1 1 0.74 (0.51, 1.1) 3 5.1 (2.8, 9.4) 27 4.7 (2.3, 9.4) 19 2 0.52 (0.33, 0.80) 9 13 (6.9, 23) 67 27 (13, 56) 77 3 0.26 (0.17, 0.41) 34 36 (20, 66) 136 113 (53, 239) 152 Odds ratios are adjusted for all variables that remained significant (p<0.05) in multivariable analysis.

2 represents the contribution of the variable at hand in the final multivariable model. § odds ratio >1: increased preference for PCI rather than medical treatment. † odds ratio >1: increased preference for CABG rather than medical treatment. ‡ odds ratio >1: increased preference for CABG rather than PCI.


112

Patients scheduled for PCI were younger, and less often had congestive

heart failure than the remaining patients. Their anginal complaints were

usually moderate, whereas only a limited number of diseased segments

was observed. The number of total occlusions and type C lesions was low.

As compared to patients scheduled for medical treatment, patients

scheduled for PCI less often had a history of CABG or myocardial infartion.

As compared to patients scheduled for CABG, they more often had a

history of CABG, but less often had proximal LAD or left main disease.

Patients in whom the physician preferred CABG were older, more often

men, and more often had peripheral vascular disease, cerebro-vascular

disease, and valvular heart disease than the remaining patients. They often

had severe anginal symptoms, a large area of myocardium at risk and

extensive vessel disease. The number of lesions suitable for PCI was low.

Determinants of intended revascularization versus medical treatment

The number of lesions suitable for PCI was the most important determinant

of intended PCI versus medical treatment: a PCI was 28 times more likely

in patients with at least one lesion suitable for such approach (table 3). The

severity of the anginal symptoms, the amount of jeopardized myocardium

and left ventricular function were also important determinants of the choice

for PCI versus medical treatment. Patients in CCS anginal class II to IV

were 1.8 to 6.2 times more likely to undergo PCI than those in class I.

Patients with an intermediate to large amount of myocardium at risk were

2.0 to 2.4 times more likely to undergo PCI than those with a small amount.

In contrast, patients with moderately to severely impaired left ventricular

function were 2.6 to 3.7 times less likely to be scheduled for PCI (odds ratio

0.38 to 0.27).

Chapter 7 ___________________________________________________________________

113

The number of lesions suitable for CABG was the most important

determinant of intended CABG versus medical treatment: a CABG was 5.1

to 36 times more likely in patients with one to three or more lesions suitable

for the surgical approach (table 3). The severity of the anginal symptoms,

the amount of jeopardized myocardium and left ventricular function were

also important determinants of the choice for CABG versus medical

treatment. However, whether or not a CABG has been performed in the

past was a more decisive factor: patients with a history of CABG were 25 to

50 less likely to be scheduled for CABG.

Determinants of intended CABG versus PCI

In patients selected for PCI or CABG, the number of lesions suitable for

each other technique was the most dominating factor for the final treatment

choice (table 3). A history of CABG, the presence of left main disease, and

the presence of valvular heart disease were also important determinants of

the phycisians preference for either CABG or PCI. Diabetes and left

ventricular function were not associated with the phycisians choice in this

respect.

Performance of predictive models

The discriminative power of the model to separate patients with a

physician’s preference for medical versus revascularization treatment was

good. The c-statistics for the PCI versus medical treatment and CABG

versus medical treatment models were 0.85 and 0.92, respectively. The

Hosmer-Lemeshow tests for calibration were non-significant (p=0.33 and

p=0.29, respectively; figure 1, top and middel panel), indicating that the

models accurately predicted the probability for a patient to be selected for


114

coronary revascularization (PCI or CABG). The discriminative power of the

model to separate patients with a physician’s preference for CABG versus

PCI treatment was also excellent (c-statistic 0.95). The Hosmer-Lemeshow

test for calibration was significant (p<0.001), but the difference between the

expected and observed probabilities was less than 3% in 8 patient

subgroups based on the deciles of the probability distribution (figure 1,

bottom panel).

Discussion

In this analysis of patients with stable angina and angiographically

significant coronary artery disease who enrolled the EHS-CR,

revascularization was generally preferred in patients with more severe

anginal complaints, an intermediate to large area of myocardium at risk,

and a preserved left ventricular function, who had not undergone prior

coronary revascularization, provided lesions were suitable for treatment.

Coronary surgery was preferred over PCI in patients with multivessel or left

main disease, as well as in those with concomitant valvular heart disease,

provided a sufficient number of lesions were suitable for CABG.

_________

Figure 1 (opposite page): Expected versus observed proportions of patients with a physician’s preference for specific treatment within deciles of expected proportions, which are based on multivariable logistic regression models. Top panel: Expected proportion of patients with a phycisian’s preference for PCI rather than medical treatment. Middle panel: Expected proportion of patients with a phycisian’s preference for CABG rather than medical treatment. Bottom panel: expected proportion of patient with a phycisian’s preference for CABG rather than PCI.

Chapter 7 ___________________________________________________________________

115


116

To a large extent, these choices are coherent with the 1999 and 2002

guidelines. There were, however, some intriguing variations between the

guidelines and the routine practice that is reflected in this survey. Impaired

left ventricular function was associated with an increased preference for

medical treatment. This contrasts with guidelines, which recommend

coronary revascularization for most patients with multivessel disease,

especially those with abnormal left ventricular function [4]. Indeed,

randomized trials comparing CABG with medical treatment in multivessel

disease demonstrated a significant improvement of symptoms, as well as a

significant mortality reduction during long-term follow-up in favor of

revascularization treatment, regardless left ventricular function [11]. There

are only a few randomized trials comparing PCI versus medical treatment,

but guidelines speculate that patients with an intermediate to large

myocardium at risk (such as those with two-vessel disease and a proximal

LAD lesion) will profit as much from PTCA as CABG. The rationale behind

the deviation from the guidelines that we observed is not entirely clear. In

fact, the interpretation of our data may be hampered by the heterogeneous

nature of the medically treated patients. In 45% of patients selected for

medical treatment, phycisians reported a low-risk natural history as their

main motivation behind their choice (table 4). At the other hand, in 16% of

patients, medical treatment was chosen because of a suspected high risk

of peri-operative complications.

Patients with a prior history of CABG were much more often selected for

medical treatment, independent of the severity of anginal symptoms, the

extent of coronary disease, and the number of lesions suitable for PCI or

CABG. Clinical trials comparing re-CABG versus medical treatment in

stable angina are lacking, as are trials comparing re-CABG with PCI. The

Chapter 7 ___________________________________________________________________

117

fact that physicians have been educated with the principle ‘primum non

nocere’ (“first do no harm”, “better safe than sorry”), may therefore partly

explain our findings.

It is interesting that diabetes was not associated with more often preference

for CABG [17]. Possibly, since large-scale randomized trials are lacking, it

is still unclear whether optimal medical management or revascularization

should be preferred in the general population of patients with stable angina

and diabetes. The ongoing BARI 2D trial, in which diabetic patients will be

randomly allocated to aggressive medical management, targeting at

optimal glycemic and metabolic control, or revascularization, may help to

solve this ‘burgeoning dilemma’, as the investigators call it [18].

Limitations

Our study has several limitations, which should be acknowledged. First, the

EHS surveys were conducted mainly in hospital settings with liberal access

to coronary revascularization facilities. It is known that the availability of

specific medical resources, including coronary revascularization, decreases

the threshold for its use [19]. Second, patients were only enrolled the EHS-

Table 4: Reasons for choosing medical treatment only.Reason (not exclusive) Percentage of patients General condition of the patient not suitable for PCI 13 General condition of the patient not suitable for CABG 13

Low-risk natural history 45 Very high-risk procedure 16 Cardiac contra-indication 6 Vessels not suitable for PCI or CABG 49 Patient refused PCI 4 Patient refused CABG 9


118

CR once the presence of coronary disease had been established by

angiography. Obviously, there are many patients with stable angina without

recent documentation of the coronary anatomy, who were not studied.

Third, no Core Lab analysis was performed of the qualifying coronary

angiogram. Consequently, detailed anatomic information that may have

influenced treatment decisions was lacking. Finally, the present data have

been acquired prior to clinical availability of drug-eluting stents [20,21]. To

which extent the availability of these devices will introduce a change in

practice and treatment strategy will be examined in a new survey.

Therefore the current results will serve as a benchmark for the evaluation of

the impact of drug-eluting stents on the practice of revascularization.

Conclusions

The EHS-CR learned that treatment decisions in patients with stable CAD

are largely in agreement with current guidelines. Most important deviations

between guideline recommendations and clinical practice patterns were

seen in patients with extensive coronary disease, impaired left ventricular

function and those with diabetes. The EHS-CR also revealed that treatment

decisions in patients with stable CAD are determined by multiple factors.

Interestingly, several of these factors were exclusion criteria in clinical trials

that have been conducted in these patients, including prior

revascularization, left main disease, renal failure and heart failure. Thus, in

order to improve evidence-based medicine, guidelines would need to

account for this and incorporate specific recommendations regarding the

best care for these specific patient subgroups.

Chapter 7 ___________________________________________________________________

119

References

1. Management of stable angina pectoris. Recommendations of the Task Force of the European Society of Cardiology. Eur Heart J 1997;18:394-413. 2. Gibbons RJ, Chatterjee K, Daley J, Douglas JS et al. ACC/AHA/ACP-ASIM Guidelines for management of patients with chronic stable angina: executive summary and recommendations. Circulation 1999; 99:2829-2848 3. Gibbons RJ, Abrams J, Chatterjee K, Daley J, Deedwania PC, Douglas JS, et al. ACC/AHA 2002 guideline update for the management of patients with chronic stable angina summary article: a report of the American College of Cardiology/American Heart Association Task Force on practice guidelines (Committee on the Management of Patients With Chronic Stable Angina). www.acc.org/clinical/guidelines/stable/stable.pdf4. Eagle KA, Guyton RA, Davidoff R, Edwards FH, Ewy GA, Gardner TJ, et al. ACC/AHA 2004 Guideline Update for Coronary Artery Bypass Graft Surgery: Summary Article: A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Update the 1999 Guidelines for Coronary Artery Bypass Graft Surgery). Circulation 2004;110:1168-1176. 5. Silber S, Albertsson P, Aviles FF, Camici PG, Colombo A, Hamm C, et al. Guidelines for percutaneous coronary interventions: the task force for percutaneous coronary interventions of the European Society of Cardiology. Eur Heart J 2005;26:804-847. 6. Abrahams J. Chronic stable angina. N Engl J Med 2005;352:2524-33 7. Yusuf S, Zucker D, Peduzzi P, Fisher LD, Takaro T, Kennedy JW, et al. Effect of coronary artery bypass graft surgery on survival: overview of 10-year results from randomised trials by the Coronary Artery Bypass Graft Surgery Trialists Collaboration. Lancet 1994;344:563-570. 8. Hoffman SN, Tenbrook JA, Wolf M et al. A Meta-analysis of randomized controlled trials comparing Coronary Artery Bypass Graft with Percutaneous Transluminal Coronary Angioplasty: One to eight year outcomes. J Am coll Cardiol 2003;41:1293-304. 9. Henderson RA, Pocock SJ, Clayton TC et al. Seven year outcome in the RITA 2 trial: coronary Angioplasty versus medical treatment. J Am Coll Cardiol 2003;42:1161-70. 10. Charanjit S. Rihal, Dominic L. Raco. Bernard J Gersh. Indications for coronary artery bypass surgery and percutaneous coronary intervention in chronic stable angina. Circulation 2003;108:2439-2445. 11. Detre KM, Holobkov R. Coronary revscularisation on Balance: Robert L Frye Lecture.Mayo Clin Proc 2002;77:72-82. 12. http://www.escardio.org/knowledge/ehs.13. Lenzen MJ, Boersma E, Bertrand ME, Maier W, Moris C, Piscione F, et al. Management and outcome of patients with established coronary artery


120

disease: the Euro Heart Survey on coronary revascularization. Eur Heart J 2005;26:1169-1179. 14. Akaike H. Information theory as an extension of the maximum likelihood principle. In: Petrov BN, Csaki F, eds. Second International Symposium on Information Theory. Budapest, Hungary: Akademia Kiado, 1973:267-281. 15. Steyerberg EW, Harrell FE Jr., Borsboom GJJM, Eijkemans MJC, Vergouwe Y, Habbema JDF. Internal validation of predictive models: Efficiency of some procedures for logistic regression analysis. J Clin Epidemiol 2001;54:774-81. 16. Hosmer DW, Lemeshow S. Applied Logistic Regression. 1989, New York: John Wiley & Sons, Inc. 17. Breeman A, Bertrand ME, Ottervanger JP, Hoeks S, Lenzen M, Sechtem U,

Legrand V, Boer de MJ, Wijns W, Boersma E. Diabetes does not influence treatment decisions regarding revascularisation in patients with stable coronary artery disease. Submitted.

18. Sobel BE, Frye R, Detre KM. Burgeoning dilemmas in the management of diabetes and cardiovascular disease: rationale for the Bypass Angioplasty Revascularization Investigation 2 Diabetes (BARI 2D) Trial. Circulation 2003;107:636-642. 19. Stenestrand U, Wallentin L. Early revascularisation and 1-year survival in 14-

day survivors of acute myocardial infarction: a prospective cohort study. Lancet 2002; 359:1805-1811.

20. Moses JW, Leon MB, Popma JJ et al. For the Sirius Investigators. Sirolimus- Eluting stents versus standard stents in patients with stenosis in a native coronary artery. N Engl J Med;349:1315-23. 21. Stones GW, Ellis SG, Cox DA et al. For the Taxus IV investigators A Polymer- based, paclitaxel-eluting stent in patients with coronary artery disease. N Engl J Med 2004;350:221-31. 22. Nashef SA, Roques F, Michel P, Gauducheau E, Lemeshow S, Salamon R. European system for cardiac operative risk evaluation (EuroSCORE). Eur J Cardiothorac Surg 1999;16:9-13.

Chapter 7 ___________________________________________________________________

121


122

Appendix Organisation of the survey

Survey Expert Committee

W. Wijns (Survey Chairman), Belgium; N. Mercado (Research Fellow), The

Netherlands; M. Bertrand, France; W. Maier, Switzerland; B. Meier,

Switzerland; C. Moris, Spain; F. Piscione, Italy; U. Sechtem, Germany; P.

Sergeant, Belgium; E. Stahle, Sweden; J. Vos, The Netherlands; P.

Widimsky, Czech Republic; F. Unger, Austria.

Euro Heart Survey Team (European Heart House, France)

Malika Manini (Operations Manager); Claire Bramley (Data Monitor);

Valérie Laforest (Data Monitor); Charles Taylor (Database Administrator);

Susan Del Gaiso (Administrator).

National Coordinators

Austria, Kurt Huber; Belgium, Guy De Backer; Bulgaria, Vera Sirakova;

Czech Republic, Roman Cerbak; Denmark, Per Thayssen; Finland, Seppo

Lehto; France, Jean-Jacques Blanc, François Delahaye; Georgia, Bondo

Kobulia; Germany, Uwe Zeymer; Greece, Dennis Cokkinos; Hungary,

Kristof Karlocai; Ireland, Ian Graham, Emer Shelley; Israel, Shlomo Behar;

Italy, Aldo Maggioni; Lithuania, Virginija Grabauskiene; The Netherlands,

Jaap Deckers; Norway, Inger Asmussen; Poland, Janina Stepinska;

Chapter 7 ___________________________________________________________________

123

Portugal, Lino Gonçalves; Russia, Vyacheslav Mareev; Slovakia, Igor

Riecansky; Slovenia, Miran F. Kenda; Spain, Angeles Alonso, José Luis

Lopez-Sendon; Sweden, Annika Rosengren; Switzerland, Peter Buser;

Turkey, Tugrul Okay; Ukraine, Oleg Sychov; United Kingdom, Kevin Fox.

Euro Heart Survey Board Committee

David Wood (Chairman), United Kingdom; Angeles Alonso, Spain; Shlomo

Behar, Israel; Eric Boersma, The Netherlands; Harry Crijns, The

Netherlands; Kim Fox, United Kingdom; Malika Manini, France; Keith

McGregor, France; Barbara Mulder, The Netherlands; Sylvia Priori, Italy;

Lars Rydén, Sweden; Luigi Tavazzi, Italy; Alec Vahanian, France; Panos

Vardas, Greece; William Wijns, Belgium; Uwe Zeymer, Germany.

Industry Sponsor

Eucomed.

Sponsoring Institutions

French Federation of Cardiology, Hellenic Cardiological Society,

Netherland Heart Foundation, Swedish Heart and Lung Foundation, and

individual hospitals.

CHAPTER 8

Diabetes does not influence treatmentdecisions regarding revascularization

in patients with stable coronary artery disease

A. Breeman, M.E. Bertrand, J.P. Ottervanger, S. Hoeks

M. Lenzen, U. Sechtem, V. Legrand, M.J. de Boer

W. Wijns, E. Boersma

Submitted

Chapter 8 ___________________________________________________________________

127

Abstract

Objective: To evaluate whether in stable angina, preference for coronary

revascularization by either percutaneous coronary intervention (PCI) or

coronary bypass surgery (CABG) is influenced by diabetic status and

whether this has prognostic implications.

Research Design and Methods: 2928 consecutive patients with stable

angina who were enrolled in the prospective Euro Heart Survey on

Coronary Revascularization were studied. Multivariable analyses were

applied to evaluate the relation between diabetes, treatment decision and

one-year outcome.

Results: Diabetes was documented in 587 patients (20%), who had more

extensive coronary disease. Revascularization was intended in 74% of

patients with diabetes and in 77% of those without diabetes. In patients

selected for revascularization, CABG was intended in 35% of diabetic and

in 33% of non-diabetic patients. Multivariable analyses did not change

these findings, but in some subgroups diabetes influenced treatment

decisions. For example, diabetics with mild heart failure had more often

intended revascularization (91%) than those without diabetes (67%,

p<0.001). Treatment decisions in patients with more extensive (left main,

multivessel, or proximal LAD) disease were not influenced by diabetic

status. Diabetes was not associated with an increased incidence of all-

cause death, non-fatal CVA or non-fatal MI at one-year, regardless

preferred treatment. The incidence of the combined end points was 7.3% in

diabetic and 6.8% in non-diabetic patients (adjusted hazard ratio 1.0 and

95%CI 0.7 to 1.4).

Treatment decisions in stable angina: diabetes ___________________________________________________________________

128

Conclusion: In stable angina, treatment decisions regarding

revascularization or the choice for CABG or PCI were not influenced by

presence of diabetes. Diabetes was not associated with a poor prognosis.

Introduction

Prevalence of diabetes in the general population is high, particularly in

North America (7.9%) and in Europe (7.8%) [1]. In patients with established

coronary artery disease (CAD) the prevalence of diabetes is even higher. In

the Euro Heart Survey on Diabetes and the Heart 14% of patients with

stable CAD had newly detected diabetes, whereas around 37% had

impaired glucose regulation. Furthermore, patients with CAD and diabetes

may have a worse prognosis [2,3].

Whether coronary revascularization in patients with diabetes has

comparable benefits as in patients without diabetes is not yet clear. There

is also debate whether coronary artery bypass surgery (CABG) or

percutaneous coronary intervention (PCI) should be preferred in diabetes

[4-6]. As a consequence, current treatment guidelines do not provide a firm

treatment advise for the general diabetic CAD population, although some

detailed recommendations are given for specific subgroups [7-14].

The Euro Heart Survey on Coronary Revascularization (EHS-CR) was

developed to obtain quantitative information on the adherence to guidelines

and prognosis in patients undergoing coronary angiography [15]. The EHS-

CR enrolled 3006 consecutive patients with stable CAD, and this well-

characterised study population provides a unique opportunity for a

Chapter 8 ___________________________________________________________________

129

systematic analysis of the relation between patient characteristics

(including diabetes), invasive treatment choices and prognosis.

Methods

Euro Heart survey of Coronary Revascularization

The EHS-CR was described in detail elsewhere [15]. Briefly, the survey

was designed to screen consecutive patients undergoing invasive

procedures in the catheterisation laboratory. Patients were enrolled if they

had a diameter stenosis of at least 50% in at least one major epicardial

coronary artery. Data were collected by dedicated data collecting officers

and sent to a central database in the European Heart House (Sophia

Antipolis, France) via the www, using the MacroTM software (InferMed,

UK). The collected data included demographics, co-morbidity, diagnosis,

and detailed information regarding diagnostic angiography and treatment

modalities. Between November 1, 2001 and March 1, 2002 a total of 5767

patients were enrolled.

Treatment decisions

The EHS-CR is a descriptive study, and the survey protocol did not dictate

any treatment decision. In general, physicians were encouraged to treat

their patients in conformance with the most recent guidelines. In order to be

informed of the physicians preferred, intended treatment, the survey

included the question: “As the treating physician, which treatment option

would be your first choice? “


130

According to the reply on this question, patients were classified with a

physician’s intention for medical treatment, PCI, or CABG.

Definitions

Since the EHS-CR was a survey of day-to-day clinical practice, it was

avoided to require additional specific diagnostic tests. In this context, the

survey protocol did not require specific measurements to verify the

diagnosis of co-morbidities, including diabetes. For this study, patients with

diabetes were classified as those who used oral hypoglycaemic agents,

insulin or the combination. The extent of CAD was estimated by the number

of diseased arteries and the number of diseased segments [15].

Follow-up

Patients were followed for one year after the initial angiography. However,

because of logistic reasons, 14 of the 130 hospitals (11%) were not able to

provide long-term follow-up information. Consequently, follow-up duration

was only 30 days in 8% of patients and 30-300 days in another 13%. The

median follow-up duration was 356 days (interquartile range: 308-365

days). Follow-up data included all-cause mortality, cerebro-vascular

accidents (CVAs), myocardial infarctions (MIs), and repeat

revascularization procedures.

Data description and data analysis

Continuous data are described as median values and corresponding

quartiles, dichotomous data as counts and percentages. Univariable

analyses were performed by unpaired Mann-Whitney tests (continuous

Chapter 8 ___________________________________________________________________

131

data) and Chi-square or Fisher’s exact tests (dichotomous data) as

appropriate.

A number of multivariable logistic regression analyses were applied to

further evaluate the relation between diabetic status and intended treatment

(medical versus revascularization; CABG versus PCI in patients with

intended revascularization), and the extent to which this relation was

influenced by a range of clinical and angiographic characteristics.

Therefore, each separate regression model included a diabetic status

interaction term. If there was statistical evidence that this interaction term

contributed to the model, it was concluded that the relation between

diabetes and intended treatment was influenced by that specific

characteristic.

As suggested by previous trials [13], we studied the relation between

diabetic status and intended treatment in three specific patient subgroups in

more detail: patients with significant left main disease; patients with 3-

vessel disease, combined with those with 2-vessel disease and a

significant lesion in the proximal LAD; and patients with 2-vessel disease

not involving the proximal LAD, combined with those with 1-vessel disease.

Kaplan-Meier survival analyses, as well as univariable and multivariable

Cox’ proportional hazard regression were applied to study patient outcome

at one-year follow-up. Since the number of outcome events was limited, in

multivariable analyses we only adjusted for the EuroSCORE [16]. We

report adjusted hazard ratios and corresponding 95% confidence intervals

(CI).

All statistical tests were 2-sided, and significance was stated at the

classical 0.05 probability level.


132

Table 1: Characteristics of the study population according to diabetic status.

Patients with diabetes (N=587)

Patientswithout diabetes (N=2341)

P

Demographics Age, years 64 (57, 71) 64 (55, 71) Men 70 78 ‡

Medical history and medication at admission Prior PCI 25 24 Prior CABG 15 12 Prior myocardial infarction 43 43 Peripheral vascular disease 18 12 ‡ Cerebro-vascular disease 10 7 § Chronic renal insufficiency 5 3 §

Use of beta-blockers, calcium antagonists, or nitrates None 9 11 Mono 34 37 Double 46 40 Triple 11 12

Presentation Concomitant valvular heart disease 7 7 NYHA symptomatic (heart failure) class I or II 9 10 III or IV 7 5 CCS angina class I 14 16 II 50 47 III 31 32 IV 5 5 Size of the myocardium at risk * ‡ small 23 29 intermediate 53 53 large 24 18 EuroSCORE ¶ 4 (2, 5) 3 (2, 5) ‡ Left ventricular function # † mild impairment (LVEF 40-50%) 26 22 moderate impairment (LVEF 30-40%) 9 7 severe impairment (LVEF <30%) 4 4 Angiographic characteristics Mitral insufficiency potentially requiring surgery 13 14 Proximal LAD disease 35 32

Chapter 8 ___________________________________________________________________

133

Table 1: continued Patientswith diabetes (N=587)

Patientswithout diabetes (N=2341)

P

Number of diseased arteries ‡ 1 26 36 2 27 30 3 38 25 left main disease 9 9 Number of diseased segments ‡ 1 22 31 2 19 24 3 20 17

4 38 28 Grafts with >50% diameter stenosis 60 61

Continuous data (age, EuroSCORE) are presented as median values (25th - 75th percentile); dichotomous data are presented as percentages; P values: § <0.05; † <0.01; ‡ <0.001. * Qualitative estimate based on non-invasive diagnostics as described in the ACC/AHA

guidelines [10]. ¶ European System for Cardiac Operative Risk Evaluation, which is a score developed to quantify

the risk of perioperative mortality in patients scheduled for cardiac surgery [16]. # Based on quantitative or qualitative measurements. In patients with a history of prior CABG only.

Results

Patient characteristics

A total of 3006 patients were diagnosed with stable angina. Complete data

on diabetic status and intended treatment was available in 2928 (97%)

patients, who compose our study population. A total of 587 (20%) patients

had diabetes.

There were important differences in clinical and angiographic baseline

characteristics between patients with and without diabetes (table 1).

Patients with diabetes were more often women, and had a higher

prevalence of chronic renal insufficiency, peripheral vascular disease and

cerebro-vascular disease. The size of the myocardium at risk was


134

significantly increased in diabetics, as was the Euro-SCORE and the

number of diseased vessels and segments. Moreover, patients with

diabetes more often had impaired left ventricular function than those

without diabetes.

Univariable analyses

In patients with diabetes, a PCI was intended in 282 (48%) patients and

CABG in 155 (26%) patients, compared to 1217 (52%) and 586 (25%) in

patients without diabetes respectively. There were, however, some specific

patient subgroups in which the presence of diabetes seemed to influence

choice of treatment (table 2). The presence of diabetes was associated with

an increased physician’s preference for medical treatment in women, in

patients without heart failure, in those with mitral valve insufficiency, and in

patients with 4 or more diseased segments. In contrast, diabetes was

associated with an increased preference for revascularization in patients

with previous PCI, and those with mild heart failure. In the diabetics, an

increased physician’s preference for CABG rather than PCI was observed

in patients below the age of 60 years, in women, in patients with previous

PCI or extensive anti-anginal medication, and in those with a Euro-SCORE

<3 points. An increased physician’s preference for intervention by PCI

rather than CABG was observed in diabetic patients with a large area of

jeopardised myocardium.

The extent of CAD was strongly associated with the physician’s preference

for revascularization (table 2). Diabetes did not influence this association.

Among patients selected for revascularization, CABG was preferred in

those with more extensive disease, independent of the coexistence of

diabetes.

Chapter 8 ___________________________________________________________________

135

Table 2: Intended invasive versus non-invasive treatment, and intended CABG versus PCI according to diabetic status in relation to clinical baseline and angiographic characteristics.

Characteristic No. of patients

Percentage with intended

revascularization

Percentage with intended CABG, in those

with intended revascularization

Diabetes No diabetes

P Diabetes No diabetes

P

All 2928 74 77 35 33 Age <60 yr 1165 77 79 34 26 § 60-69 yr 957 74 75 32 37

70 yr 806 72 77 42 38 Sex women 684 68 77 § 35 27 § men 2242 77 77 36 34 Prior PCI no 2218 72 79 † 43 37 § yes 697 81 71 § 15 17 Prior CABG no 2548 77 80 39 34 yes 366 59 53 10 13

no 1673 76 80 35 33 Prior myocardial infarction yes 1240 73 73 36 31

no 2535 74 77 33 31 Peripheral vascular disease yes 376 75 76 48 41

no 2685 75 77 34 32 Cerebrovascular disease yes 225 67 78 53 41

no 2822 75 77 36 33 Chronic renal insufficiency yes 102 60 72 22 23

none 314 64 73 32 37 mono 1060 74 74 34 29 double 1209 76 80 34 35

Use of betablockers, calciumantagonists, nitrates prior to inclusion

triple 332 77 80 47 32 §

no 2717 75 77 34 30 Concomitant valvular heart disease yes 209 63 77 62 62

no heart failure

2469 74 79 § 33 30

I or II 299 91 67 ‡ 47 46

NYHA symptomatic (heart failure) class

III or IV 159 60 67 48 53

I 437 69 61 29 26 II 1353 72 77 33 29 III 903 79 82 42 40

CCS angina class

IV 131 85 92 35 39 Table 2 continues on page 136


136

Table 2 continued No. of patients

Percentage with intended

revascularization

Percentage with intended CABG, in those

with intended revascularization

Diabetes No diabetes

P Diabetes No diabetes

P

small 707 64 65 18 20 Inter-mediate

1324 75 80 40 31 Size of the myocardium at risk

large 481 80 78 32 46 §

<3 * 1129 80 82 32 23 § EuroSCORE 3 1760 72 74 37 38

>50% 1694 75 80 § 34 31 41-50% 597 77 76 45 38 31-40% 184 71 64 43 52

Left ventricular ejection fraction

30% 98 57 51 23 53 Mitral valve insufficiency no 2409 74 78 § 35 31 yes 392 76 72 44 44 Prioximal LAD disease no 1973 72 75 25 24 yes 955 79 82 53 48

1 987 72 73 8 8 2 863 77 80 18 24 3 798 74 79 59 60

Number of diseased arteries

Left Main 272 75 80 67 74 1 861 74 73 7 10 2 667 78 79 18 21 3 512 74 78 46 37

Number of diseased segments

>3 888 73 80 § 55 62 0 807 49 53 84 79 1 1073 85 86 14 13 2 590 85 86 27 20

Number of diseased segments suitable for PCI

>2 458 85 85 40 46 0 701 57 60 5 8 1 755 83 80 10 10 2 511 76 83 28 28

Number of diseased segments suitable for CABG

>2 961 81 84 67 67 0 144 44 41 7 13 1 110 62 57 6 10

Grafts with > 50% diameter stenosis

>2 112 76 66 16 16

P-value: § <0.05; † <0.01; ‡ <0.001. * Median value.

Chapter 8 ___________________________________________________________________

137

Multivariable analyses

After multivariable adjustment for a range of potential cofounders (table 3),

the odds ratio for the relation between diabetic status and preferred

revascularization versus medical treatment was 0.91 (95% CI 0.70 to 1.17).

Thus, the treating physicians did not have a differential preference for either

medical treatment or coronary revascularization in relation to diabetic

status. There was also no preference for either CABG or PCI in relation to

diabetic status, with the adjusted odds ratio 0.92 (95% CI 0.63 to 1.3).

However, diabetes significantly influenced treatment decisions in several

subgroups according to age, previous PCI, heart-failure, concomitant

valvular disease and Euro-Score. For example, the presence of diabetes

was associated with an increased preference for coronary revascularization

in patients with mild heart failure (NYHA class I or II). Furthermore,

diabetes was associated with an increased preference for CABG rather

than PCI in patients below the age of 60 years, but associated with an

increased preference for PCI in elderly patients.

Outcome after one-year

The incidence of the composite endpoint of all-cause death, non-fatal CVA

or non-fatal MI at one-year follow-up was 7.3% in patients with diabetes

and 6.8% in patients without diabetes (adjusted hazard ratio 1.0 and 95%CI

0.7 to 1.4). Patients with more extensive disease had a higher incidence of

death, CVA or MI at one-year follow-up than those with less extensive

disease (figure 1, table 3). However, importantly, in the subgroups

according to the extent of CAD there were no significant differences in the

incidence of this composite endpoint between patients with and without

diabetes.


138

Figu

re 1

: In

cide

nce

of a

dver

se c

ardi

ac e

vent

s du

ring

1 ye

ar f

ollo

w-u

p ac

cord

ing

to t

he e

xten

t of

cor

onar

y di

seas

e an

d di

abet

ic s

tatu

s. U

pper

pan

el p

rese

nts

the

inci

denc

e of

dea

th,

cere

bro-

vas

cula

r ac

cide

nts

and

myo

card

ial i

nfar

ctio

n. L

ower

pa

nel

pres

ents

th

e in

cide

nce

of

deat

h,

cere

bro-

va

scul

ar

acci

dent

s,

myo

card

ial

infa

rctio

n,

and

repe

ated

co

rona

ry

reva

scul

ariz

atio

n pr

oced

ures

. The

bol

d lin

e re

pres

ent p

atie

nts

with

dia

bete

s.

Chapter 8 ___________________________________________________________________

139

Table 3: Multivariably adjusted association between treated diabetes and intended treatment according to patient characteristics with significant interaction.

Intended revascularization (versus medical treatment)

Intended CABG (versus PCI) §

All Adjusted odds ratio (95% CI)*

P-value for homogenity

Adjusted odds ratio (95% CI)*

P-value for homogeneity

All 0.91 (0.70, 1.17) 0.92 (0.63, 1.3) Age <60 yr 1.6 (0.88, 2.9) reference 60-69 yr 0.53 (0.22, 1.3) 0.004

70 yr 0.63 (0.25, 1.5) 0.29 Prior PCI no 0.74 (0.55, 1.0) reference yes 1.7 (0.93, 3.1) 0.007

no heart failure

0.77 (0.59, 1.0) reference

I or II 8.3 (2.8, 25) <0.001 III or IV 0.63 (0.25, 1.5) 0.85

no 1.0 (0.7, 1.5) reference Concomitant

valvular heart disease

yes 0.25 (0.06, 0.95)

0.047

<3 2.1 (1.1, 3.7) reference EuroSCORE3 0.55 (0.26, 1.2) <0.001

* An odds ratio>1 implies that treated diabetes is associated with an increased preference for the treatment option in the corresponding column, whereas an odds ratio <1 implies that treated diabetes is associated with an increased preference for the alternative. Odds ratios are adjusted for all variables that are listed in tables 2 and 3, as appropriate. § The analysis is limited to patients with intended coronary revascularization.

Concerning the relation between diabetic status, treatment choice and

outcome, because of small numbers, patients with left main disease or 3-

vessel disease were considered as one group. Regardless of the extent of

the disease, patients selected for PCI had a lower incidence of major

adverse cardiovascular or cerebro-vascular events at 1-year follow-up than

those selected for medical treatment or CABG (table 4), but again we

observed no difference in the relation between intended treatment and

outcome according to diabetic status.


140

Table 4: Incidence of adverse cardiac complications at 1 year follow-up according to the extent of coronary disease, intended treatment and diabetic status.

Intended treatment

Diabetes No. of patients

Incidenceof death,

CVA or MI at 1 year§

Adjustedhazard ratio(95% CI)†

Incidence of death, CVA,

MI, (re)CABG or (re)PCI at

1 year ‡

Adjustedhazard ratio (95% CI)†

Any Yes 587 7.3 1.0 (0.7, 1.4) 15.9 1.0 (0.8, 1.3) No 2341 6.8 1 13.8 1 Medical Yes 150 6.7 0.9 (0.5, 1.7) No 538 7.6 1 PCI Yes 282 5.3 1.0 (0.6, 1.8) 16.8 1.1 (0.8, 1.6) No 1217 4.9 1 14.8 1 CABG Yes 155 12.0 1.1 (0.6, 1.9) 12.8 1.0 (0.6, 1.8)

All patients

No 586 10.2 1 11.7 1


Left main OR3 vessel disease OR2 vessel disease and proximal LAD

No 472 9.3 1 11.1 1


1 vessel disease OR2 vessel disease, but no proximal LAD

No 114 13.5 1 14.2 1 § Kaplan-Meier estimate. † Hazard ratio’s are adjusted for differences in EuroSCORE between groups. ‡ Kaplan-Meier estimate; analysis based on patient with intended coronary revascularization only.

Chapter 8 ___________________________________________________________________

141

Discussion

This analysis revealed that in stable angina subsequent treatment

decisions regarding revascularization (and the choice for either CABG or

PCI) were not influenced by the presence of diabetes. Importantly, diabetic

patients with left main disease, proximal LAD disease, or more extensive,

multivessel disease were not more likely to undergo coronary

revascularization (or CABG) than their non-diabetic counterparts. Diabetes

was also not associated with a poor prognosis.

Previous (1999) and current (2002) guidelines recommend CABG rather

than PCI in patients with diabetes and multivessel disease [10,11]. This

treatment advise is mainly based on post-hoc analyses from randomised

trials that were conducted before the use of stents and glycoprotein (GP)

IIb/IIIa inhibitors. In the BARI trial, patients with treated diabetes who were

randomised to CABG had significant better survival after 7-year follow-up

than those randomised to balloon angioplasty [4]. To some extent, this

finding was confirmed by a meta-analysis of 13 randomised trials of CABG

versus balloon angioplasty (including BARI), which demonstrated improved

survival in favour of CABG at 4-year follow-up, but no longer at 6.5-year

follow-up [17].

How can we understand the discrepancy between guidelines

recommendations and the clinical practice patterns that we observed?

Possibly, those involved in the care of patients with stable angina are not

convinced that patients with diabetes should be treated differently than

those without diabetes, especially since large-scale randomised trials are

lacking. The ongoing BARI 2D trial, in which diabetic patients will be

randomly allocated to aggressive medical management, targeting at


142

optimal glycemic and metabolic control, or revascularization, may help to

solve this ‘burgeoning dilemma’, as the investigators call it [18].

Furthermore, it is well-known that patients enrolled in clinical trials form a

selected population, particularly in randomised trials comparing PCI and

CABG [19]. In this respect, observational studies, including the BARI

registry [5], reported similar outcome after CABG and PCI in patients with

diabetes and symptomatic coronary disease [6].

Another interesting aspect is emphasised by McGuire et al., who evaluated

the effects of reporting the BARI trial results (which were made public

together with a ‘clinical alert’) on decision-making [20]. It was reported that

the rapid advancement of healthcare technology was the major factor of the

lack of influence of the clinical alert and the BARI trial on current clinical

practice. More recent randomised trials comparing CABG against stenting

with liberal use of GP IIb/IIIa inhibitors, which were conducted during 1996-

2000, demonstrating more favourable results in those with diabetes [21],

may have influenced treatment decisions.

Probably, antirestenotic, drug-eluting stents (DES) will improve the results

of PCI relative to CABG in diabetic patients [22,23]. The ongoing CARDia,

FREEDOM and BARI 2D trials, which enrol patients with coronary disease

and diabetes, who are randomised to either CABG or PCI with DES and

modern antiplatelet therapy, including GP IIb/IIIa inhibitors and clopidogrel,

may shed a vivid light on the PCI-versus-CABG debate.

In our survey, diabetes was not associated with an increased incidence of

major adverse cardiac and cerebral events. Especially diabetic patients

scheduled for PCI had similar prognosis as non-diabetics. It is true that the

limited number of patients and the short duration of follow-up might have

masked significant differences, but large differences in clinical outcome can

Chapter 8 ___________________________________________________________________

143

be excluded with sufficient certainty. These observations seem to justify the

position of PCI as a safe treatment for patients with diabetes and stable

CAD.

Limitations

This study has several limitations, which need to be addressed. First, The

EHS-CR was conducted mainly in hospitals with liberal access to coronary

revascularization facilities. It is known that the availability of specific

medical resources decreases the threshold for it’s use [24]. Thus, the

percentage of patients undergoing revascularization, particularly CABG,

might have been relative high. However, there is no indication that this has

been differently so in patients with and without diabetes.

Second, as the EHS-CR is a survey of routine clinical practice, the treating

physician was not enforced to use specific laboratory tests in order to

establish the diagnosis ‘diabetes’. Thus, we cannot exclude the possibility

that misclassification of diabetes had occurred. However, the clinical and

angiographic characteristics of patients with diabetes corresponded quite

well with other datasets of diabetics with stable coronary disease [25].

Third, no Core Lab analysis was performed of the qualifying coronary

angiography. Consequently, detailed anatomic information that may have

influenced treatment decisions was lacking. This is especially relevant for

the 366 patients with a prior history of CABG. However, sensitivity analyses

that excluded these patients showed consistent results.

Conclusions

Diabetes is not among the factors that determine treatment decisions

regarding revascularization in patients with stable coronary disease.


144

Diabetes was not associated with a worse prognosis, independent of

invasive treatment preference. In expectation of the results of ongoing

clinical trials, comparing several medical and more invasive treatment

strategies, guidelines for the management of CAD patients with diabetes

should be updated more systematically than is currently the case.

Chapter 8 ___________________________________________________________________

145

References

1. http:/www.idf.org/home. 2. Coutinho M, Gerstein HC, Wang Y, Yusuf S. The relationship between glucose and incident cardiovascular events. A metaregression analysis published data from 20 studies of 95,783 individuals followed for 12.4 years. Diabetes Care 1999;22:233-240. 3. Bartnik M, Ryden L, Ferrari R, Malmberg K, Pyorala K, Simoons M, Standl E, Soler-Soler J, Ohrvik J. The prevalence of abnormal glucose regulation in patients with coronary artery disease across Europe. The Euro Heart Survey on Diabetes and the Heart. Eur Heart J 2004;25:1880-1890. 4. The BARI Investigators. Seven-year outcome in the Bypass Angioplasty Revascularization Investigation (BARI) by treatment and diabetic status. J Am Coll Cardiol 2000;35:1122-1129. 5. Feit F, Brooks MM, Sopko G, Keller NM, Rosen A, Krone R, Berger PB, Shemin R, Attubato MJ, Williams DO, Frye R, Detre KM. Long-term clinical outcome in the Bypass Angioplasty Revascularization Investigation Registry: comparison with the randomized trial. BARI Investigators. Circulation 2000;101:2795-2802. 6. Barsness GW, Peterson ED, Ohman EM, Nelson CL, DeLong ER, Reves JG, Smith PK, Anderson RD, Jones RH, Mark DB, Califf RM. Relationship between diabetes mellitus and long-term survival after coronary bypass and angioplasty. Circulation 1997;96:2551-2556. 7. Jonathan Abrahams. Chronic stable angina. N Engl J Med 2005;352:2524- 2533. 8. Management of stable angina pectoris. Recommendations of the Task Force of the European Society of Cardiology. Eur Heart J 1997;18:394-413. 9. Silber S, Albertsson P, Aviles FF, Camici PG, Colombo A, Hamm C, Jorgensen E, Marco J, Nordrehaug JE, Ruzyllo W, Urban P, Stone GW, Wijns W. Guidelines for percutaneous coronary interventions. The Task Force for Percutaneous Coronary Interventions of the European Society of Cardiology. Eur Heart J 2005;26:804-847. 10. Gibbons RJ, Chatterjee K, Daley J, ACC/AHA/ACP-ASIM guidelines for the management of patients with chronic stable angina: executive summary and recommendations. A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on Management of Patients with Chronic Stable Angina). Circulation 999;99:2829- 2848. 11. Gibbons RJ, Abrams J, Chatterjee K, American College of Cardiology/ American Heart Association Task Force on Practice Guidelines. Committee on the Management of Patients With Chronic Stable Angina. ACC/AHA 2002 guideline update for the management of patients with chronic stable angina summary article: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on the


146

Management of Patients With Chronic Stable Angina). Circulation 2003;107:149-158. 12. Eagle KA, Guyton RA, Davidoff R, American College of Cardiology; American Heart Association Task Force on Practice Guidelines; American Society for Thoracic Surgery and the Society of Thoracic Surgeons. ACC/AHA 2004 guideline update for coronary artery bypass graft surgery: summary article: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Update the 1999 Guidelines for Coronary Artery Bypass Graft Surgery). Circulation 2004;110:1168-1176. 13. Flaherty JD, Davidson CJ. Diabetes and coronary revascularization. JAMA 2005;293:1501-1508. 14. Rihal CS, Raco DL, Gersh BJ, Yusuf S. Indications for coronary artery bypass surgery and percutaneous coronary intervention in chronic stable angina: review of the evidence and methodological considerations. Circulation 2003;108:2439-2445.15. 15. Lenzen MJ, Boersma E, Bertrand ME, Maier W, Moris C, Piscione F, Sechtem U, Stahle E, Widimsky P, de Jaegere P, Scholte op Reimer WJ, Mercado N, Wijns W. Management and outcome of patients with established coronary artery disease: the Euro Heart Survey on coronary vascularization. Eur Heart J 2005;26:1169-1179. 16. Nashef SA, Roques F, Michel P, Gauducheau E, Lemeshow S, Salamon R. European system for cardiac operative risk evaluation (EuroSCORE). Eur J Cardiothorac Surg 1999;16:9-13. 17. Hoffman SN, TenBrook JA, Wolf MP, Pauker SG, Salem DN, Wong JB. A meta-analysis of randomized controlled trials comparing coronary artery bypass graft with percutaneous transluminal coronary angioplasty: one- to eight-year outcomes. J Am Coll Cardiol 2003;41:1293-1304. 18. Sobel BE, Frye R, Detre KM. Burgeoning dilemmas in the management of diabetes and cardiovascular disease: rationale for the Bypass Angioplasty Revascularization Investigation 2 Diabetes (BARI 2D) Trial. Circulation 2003;107:636-642. 19. Hordijk-Trion M, Lenzen M, Wijns W, De Jaegere P, Simoons ML, Mercado N, Boersma E. Patients enrolled in coronary intervention trials are not representative of patients in clinical practice: results from the Euro Heart Survey on Coronary Revascularizatison. Eur Heart J 2006;27:671-8. 20. McGuire DK, Anstrom KJ, Peterson ED. Influence of the Angioplasty Revascularization Investigation National Heart, Lung, and Blood Institute Diabetic Clinical Alert on practice patterns: results from the National Cardiovascular Network Database. Circulation 2003;107:1864-1870. 21. Mercado N, Wijns W, Serruys PW, Sigwart U, Flather MD,Stables RH, O'Neill WW, Rodriguez A, Lemos PA, Hueb WA, Gersh BJ, Booth J, Boersma E. One-year outcomes of coronary artery bypass graft surgery versus percutaneous coronary intervention with multiple stenting for multisystem

Chapter 8 ___________________________________________________________________

147

disease: a meta-analysis of individual patient data from randomized clinical trials. J Thorac Cardiovasc Surg 2005;130:512-519. 22. Moses JW, Leon MB, Popma JJ, Fitzgerald PJ, Holmes DR, O'Shaughnessy C, Caputo RP, Kereiakes DJ, Williams DO, Teirstein PS, Jaeger JL, Kuntz RE; SIRIUS Investigators. Sirolimus-eluting stents versus standard stents in patients with stenosis in a native coronary artery. N Engl J Med 2003;349:1315-1323. 23. Stone GW, Ellis SG, Cox DA, Hermiller J, O'Shaughnessy C, Mann JT, Turco M, Caputo R, Bergin P, Greenberg J, Popma JJ, Russell ME; TAXUS-IV Investigators. A polymer-based, paclitaxel-eluting stent inpatients with coronary artery disease. N Engl J Med 2004;350:221-231. 24. Stenestrand U, Wallentin L. Early revascularisation and 1-year survival in 14- day survivors of acute myocardial infarction: a prospective cohort study. Lancet 2002;359:1805-1811. 25. Niles NW, McGrath PD, Malenka D, Quinton H, Wennberg D, Shubrooks SJ, Tryzelaar JF, Clough R, Hearne MJ, Hernandez F Jr, Watkins MW, O'Connor GT; 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;37:1008-1015.


148

Appendix Organisation of the survey

Survey Expert Committee

W. Wijns (Survey Chairman), Belgium; N. Mercado (Research Fellow), The Netherlands; M. Bertrand, France; W. Maier, Switzerland; B. Meier, Switzerland; C. Moris, Spain; F. Piscione, Italy; U. Sechtem, Germany; P. Sergeant, Belgium; E. Stahle, Sweden; J. Vos, The Netherlands; P. Widimsky, Czech Republic; F. Unger, Austria.

Euro Heart Survey Team (European Heart House, France)

Malika Manini (Operations Manager); Claire Bramley (Data Monitor); Valérie Laforest (Data Monitor); Charles Taylor (Database Administrator); Susan Del Gaiso (Administrator).

Industry Sponsor

Eucomed.

Sponsoring Institutions

French Federation of Cardiology, Hellenic Cardiological Society, Netherlands Heart Foundation, Swedish Heart and Lung Foundation, and individual hospitals.

Summary

and

Conclusions

Summary and Conclusions ___________________________________________________________________

151

The main objective of the studies presented in this thesis was to provide

more insight in several aspects of coronary revascularisation, including

decision-making in daily practice and (long-term) prognostic implications.

Particular attention was paid to patients with coronary artery disease (CAD)

who have multivessel disease and patients with diabetes.

Main findings of the studies

Chapter 1 provides a general introduction and describes the outline of the

thesis. Coronary revascularisation by either CABG or PCI are nowadays

frequently used and widely accepted for the treatment of several clinical

manifestations of CAD. Although both methods of coronary

revascularisation have been improved in the past decades, treatment

decision-making (medical treatment versus revascularisation, PCI versus

CABG) in individual patients may be difficult, in part because most

randomised controlled trials have low external validity.

In chapter 2 current evidence of the prognostic implication of

revascularisation in patients with Non-ST Elevation Acute Coronary

Syndrome (NSTE-ACS) is reviewed. Immediate as well as early

revascularisation may be beneficial in these patients, but has traditionally

been employed as an ‘ischemia-guided’ strategy. A number of randomised

trials (including more than 10,000 patients) and observational studies have

compared routine invasive versus selective invasive strategies. Most


152

randomised trials are limited by a high cross-over rate, whereas

observational studies are limited by selection bias. Data from registries

demonstrated benefits with revascularisation, with lower mortality. Results

from randomised trials are less clear with regard to mortality reduction,

although long-term survival after hospital discharge may be better after the

invasive approach. In the randomised trials there was a decreased risk of

death or MI after the invasive approach (OR 0.88, 95% CI 0.76-1.0). It can

be concluded that the optimal timing of angiography and subsequent

revascularisation is still somewhat uncertain, but that angiography and

revascularisation should be considered in every patient admitted with a

NSTE-ACS, particularly in those with high-risk characteristics, such as ST

segment depression >0.1 mV, accelerated angina in the prior 2 months,

nitrate use in the prior week before admission or elevated troponin.

The objective of chapter 3 was to evaluate the relationship between the

completeness of revascularisation by percutaneous coronary intervention

and the one-year occurrence of adverse cardiac events in patients with

multivessel disease who were enrolled in the Coronary Angioplastry versus

Bypass Revascularisation Investigation (CABRI). In CABRI, patients were

randomised to CABG (n=513) or PCI (n=541). Revascularisation in patients

randomised to PCI was defined as complete if no lesions with a diameter

stenosis >50% remained post-procedure. Patients with complete

revascularisation were distinguished from those with one, two or three or

more remaining lesions, respectively. Differences in baseline

characteristics and in the one-year occurrence of death, myocardial

infarction, (re)CABG and (re)PCI between these subgroups were


153

evaluated. Comparisons were made with patients randomised to CABG.

Complete revascularisation was obtained in 148 patients randomised to

PCI (27%). In 147 (27%) cases one lesion remained, while there were 122

(23%) and 119 (22%) patients with two and three or more remaining

lesions, respectively. Five (1%) patients could not be classified. The one-

year rates of either death or MI were 9.5%, 5.4%, 8.2% and 12.6% in the

respective PCI subgroups (P=0.23), and 6.2% in patients randomised to

CABG (comparison with three or more remaining lesions after PCI: p=0.02)

The percentages of repeat interventions during one-year follow-up were

29.7%, 29.3%, 39.3% and 51.3% (p<0.001), much higher than after CABG

(3.5%; p<0.001). It was concluded that complete revascularisation by PCI

in multivessel coronary disease did not result in a lower death or MI rate

compared with incomplete revascularisation. Overall, the patient’s

prognosis after PCI is similar to CABG, but patients with three or more

remaining lesions after PCI had a worse prognosis than CABG patients,

with a higher incidence of the combined endpoint death or MI as well as

more additional procedures.

Chapter 4 describes the 6-month clinical outcome of patients with

multivessel disease enrolled in PURSUIT (Platelet Glycoprotein IIb/IIIa in

Unstable Angina: Receptor Suppression Using Integrelin Therapy). Patients

were divided into three groups according to the treatment strategy applied

during the first 30 days of enrolment. The PCI group was further

subdivided: patients receiving coronary stents were separated from those in

whom no stents were used. Mortality after 30 days was 6.7, 3.9, 2.4 and

4.8% for the medical treatment, PCI (balloon), PCI (stent) and CABG


154

groups, respectively (p value = 0.002). Differences as observed at 30 days

were still present at 6-month follow-up with 11.1, 5.8, 5.5 and 6.5%

mortality event rates for the aforementioned groups (p value = 0.002). The

30-day myocardial infarction (MI) rate was lower among medically than

non-medically treated patients, with the highest event rate observed in the

CABG group (27.7%). Approximately half of the MIs in the PCI and CABG

subgroups occurred within 48 h after the procedure. However, it was

concluded that the observed differences in clinical outcomes can be

explained by an imbalance in baseline characteristics and comorbid

conditions

In chapter 5, long-term survival is studied in unselected patients with

coronary artery disease who are potential candidates for an invasive

approach. Data from DUtch inventory of invasive Coronary Atherosclerosis

Treatments (DUCAT) study were used. All patients with significant coronary

artery disease who were presented for coronary revascularisation to two

tertiary centres in 1992 were included. Follow-up data were collected in

September 2002. A total of 877 patients were included in this analysis.

Mean age was 62 and the most common clinical diagnosis was chronic

stable angina (60%). Diabetes was present in 12% of the patients. During

the follow-up period, a total of 233 patients (27%) died. Predictors of long-

term survival were increasing age, diabetes, peripheral vascular disease

and a decreased left ventricular function. Compared to medical treated

patients, those treated with revascularisation (either by PCI or CABG) had

a decreased long-term mortality (p 0.05). Of the patients with PCI 27% had

died, compared to 24% in those who had CABG and 36% of those who


155

were treated medically. However, after adjusting for differences in baseline

variables, conservative treatment was no significant predictor of long-term

mortality. After multivariable analyses, increasing age, decreased left

ventricular function and diabetes were independent predictors of long-term

mortality.

The optimal method of coronary revascularisation in diabetic patients

remains controversial. In chapter 6, it was our aim to evaluate long-term

outcome of diabetic patients with coronary artery disease in daily practice in

whom an invasive approach was considered. As in chapter 5, data from the

DUCAT study were used. A total of 107 patients (12%) had diabetes.

Patients with diabetes were older and more frequently women. Long-term

mortality was higher in diabetes than in non-diabetes (36% vs. 25%,

p=0.01). This association was observed in both medical treated patients

(65% vs. 31%, p=0.01) and in those treated with percutaneous coronary

intervention (41% vs. 24%, p=0.02). There was, however, no difference in

mortality in diabetes versus non-diabetes after CABG (24% vs. 24%,

p=0.89). Multivariate analysis did not change these findings. It was

concluded that in patients with coronary artery disease in whom an invasive

approach is considered, diabetes was associated with a higher long-term

mortality. After coronary artery bypass grafting, long-term survival may be

comparable between patients with and without diabetes.

Determinants of clinical decision-making in patients with stable CAD were

assessed in chapter 7. The 2936 patients with stable angina pectoris who


156

enrolled the Euro Heart Survey on Coronary Revascularisation (EHS-CR)

were subject of this analysis. After the diagnosis was confirmed, physicians

decided upon treatment: only medical management, or combined by

revascularisation therapy by PCI or CABG. Logistic regression analyses

were applied to evaluate the relation between baseline characteristics and

treatment decision: medical treatment versus PCI, medical treatment

versus CABG and PCI versus CABG. Median age of the patients was 64

years, 77% were men and 20% had diabetes. Medical therapy was

intended in 690 (24%), PCI in 1503 (51%), and CABG in the remaining 743

(25%) patients respectively. Revascularisation was generally preferred in

patients with more severe anginal complaints, an intermediate to large area

of myocardium at risk, and a preserved left ventricular function, who had

not undergone prior coronary revascularisation, provided lesions were

suitable for treatment. CABG was preferred over PCI in multivessel or left

main disease, as well as in those with concomitant valvular heart disease,

provided a sufficient number of lesions were suitable for CABG. In those

with previous CABG, more often PCI was preferred than re-CABG.

Diabetes was not associated with more often preference for CABG.

This analysis of the EHS-CR revealed that treatment decisions in patients

with stable CAD are determined by multiple factors and in general are in

agreement with the guidelines. However in patients with diabetes there is

no preference of CABG over PCI. In patients with previous CABG, PCI is

the preferred therapy in the group of patients scheduled for

revascularisation.


157

The objective of chapter 8 was to evaluate whether in stable angina,

preference for coronary revascularisation by either percutaneous coronary

intervention (PCI) or coronary bypass surgery (CABG) was influenced by

diabetic status and whether this had prognostic implications. The 2928

consecutive patients with stable angina who were enrolled in the

prospective Euro Heart Survey on Coronary Revascularisation were

studied. Multivariable analyses were applied to evaluate the relation

between diabetes, treatment decision and one-year outcome. Diabetes was

documented in 587 patients (20%), who had more extensive coronary

disease. Revascularisation was intended in 74% of patients with diabetes

and in 77% of those without diabetes. In patients selected for

revascularisation, CABG was intended in 35% of diabetic and in 33% of

non-diabetic patients. Multivariable analyses did not change these findings,

but in some subgroups the presence of diabetes influenced treatment

decisions. For example, diabetics with mild heart failure had more often

intended revascularisation (91%) than those without diabetes (67%,

p<0.001). Treatment decisions in patients with more extensive (left main,

multivessel, or proximal LAD) disease were not influenced by diabetic

status. Diabetes was not associated with an increased incidence of all-

cause death, non-fatal CVA or non-fatal MI at one-year, regardless of the

preferred treatment. The incidence of the combined end points was 7.3% in

diabetic and 6.8% in non-diabetic patients (adjusted hazard ratio 1.0 and

95%CI 0.7 to 1.4). It can be concluded that in stable angina, treatment

decisions regarding revascularisation or the choice for CABG or PCI were

not influenced by presence of diabetes. Diabetes was not associated with a

poor prognosis.


158

Final comments and future directions

Coronary revascularisation by either PCI or CABG is an important

treatment option for patients with (symptomatic) coronary artery disease. It

may relief symptoms and in selected patients improve survival. Although

both revascularisation techniques have been improved during the past

decades, one of the challenges for the future is to apply the appropriate

treatment to the individual patient at the optimal moment in time. Results of

randomised controlled trials are important for making this decision.

However, most trials have also several limitations, such as a low external

validity and short follow-up period. Moreover, most randomised trials in the

field of revascularisation are not blinded, and results may be influenced by

crossover. Therefore, these trials should be combined with data from

observational studies. Because treatment regimens change, the continuous

evaluation of effects of treatment (on subgroups) in daily practice remains

mandatory.

The extent of CAD is important in the selection of the mode of

revascularisation therapy. Completeness of revascularisation may improve

prognosis. Whether in patients with multivessel disease CABG or PCI

should be the preferred strategy has to be further assessed in many

subgroups of patients with CAD, especially in those with diabetes.

Diabetes is one of the most important co-morbidities in patients with CAD

and is characterised by a rapidly increasing prevalence in the general


159

population. It is still uncertain whether PCI or CABG should be preferred in

the majority of patients with diabetes and CAD. Probably, antirestenotic,

drug-eluting stents (DES) will improve the results of PCI relative to CABG in

these patients. The ongoing CARDia, FREEDOM and BARI 2D trials, which

enroll patients with CAD and diabetes, who are randomised to either CABG

or PCI with DES and modern antiplatelet therapy, including GP IIb/IIIa

inhibitors and clopidogrel, may shed a vivid light on the PCI-versus-CABG

debate.

Samenvatting

en

Conclusies

Samenvatting en Conclusies ___________________________________________________________________

163

De studies die worden gepresenteerd in dit proefschrift hebben als

voornaamste doel om meer inzicht te verschaffen in een aantal aspecten

van coronaire revascularisatie technieken, waaronder besluitvorming in de

dagelijkse praktijk en de gevolgen daarvan op de lange termijn. Patiënten

met meervats coronairlijden of diabetes mellitus hadden onze bijzondere

belangstelling.

Belangrijkste bevindingen van de studies

Hoofdstuk 1 geeft een algemene inleiding en beschrijft de inhoud van het

proefschrift. Coronaire revascularisatie door middel van CABG of PCI wordt

tegenwoordig veelvuldig gebruikt voor de behandeling van verschillende

klinische uitingen van coronairlijden. Hoewel beide technieken gedurende

de afgelopen decennia zijn verbeterd, kan het maken van beslissingen bij

de individuele patiënt moeilijk zijn, voor een deel omdat de externe

validiteit van de meeste gerandomiseerde studies laag is.

In hoofdstuk 2 wordt een overzicht gegeven van de studies en de

registraties met betrekking tot het effect van coronaire revascularisatie op

de prognose van patiënten met een niet-ST elevatie myocardinfarct.

Vroege revascularisatie werd in het verleden vooral aanbevolen op geleide

van aantoonbare ischemie. Een aantal gerandomiseerde studies, waarin

meer dan 10.000 patiënten deelnamen, en observationele studies

vergeleken een routine invasieve benadering met een meer selectieve

invasieve strategie. De meeste gerandomiseerde studies worden beperkt


164

door een hoog “cross over” percentage, waarbij met name in de selectieve

groep, patiënten toch een revascularisatie ondergingen. Observationele

studies worden beperkt door de mogelijke verstorende invloed van selectie

bias. Deze observationele studies laten een duidelijk voordeel zien van

revascularisatie, met lagere sterfte. De resultaten van gerandomiseerde

studies zijn niet eenduidig ten aanzien van vermindering van sterfte. De

overleving op langere termijn na ontslag uit het ziekenhuis is mogelijk beter

na een invasieve aanpak. In de gerandomiseerde studies werd een afname

van de combinatie van sterfte en infarcten na een invasieve benadering

gevonden (OR 0,88, 95% CI 0,76-1,0). Er kan worden geconcludeerd dat

er nog steeds enige onzekerheid bestaat over het optimale tijdstip van

angiografie en eventuele revascularisatie, maar dat angiografie en

mogelijke revascularisatie in principe moet worden overwogen bij iedere

patiënt die zich met een niet-ST elevatie ACS presenteert, in het bijzonder

patiënten met hoog-risico karakteristieken zoals ST-segment depressie van

meer dan 0,1 mV, progressieve angina pectoris in de afgelopen 2

maanden, nitraat gebruik in de week voorafgaand aan opname of

verhoogde troponine.

In hoofdstuk 3 wordt de relatie onderzocht tussen de volledigheid van

revascularisatie door middel van PCI en het optreden van cardiale

gebeurtenissen gedurende 1 jaar bij patiënten met meervats-coronairlijden

die waren geincludeerd in de Coronary Angioplasty versus Bypass

Revascularisation Investigation (CABRI) studie. In de CABRI studie werden

patiënten gerandomiseerd naar CABG (n=513) of PCI (n=541). Van

complete revascularisatie na PCI was sprake indien er na de procedure

geen afwijkingen over waren met een diameter stenose van >50%.


165

Patiënten met complete revascularisatie werden vergeleken met patiënten

met 1, 2 en 3 of meer overgebleven afwijkingen. Verschillen in basale

kenmerken en het optreden van sterfte, myocardinfarct, (re-) CABG en (re-)

PCI tussen deze subgroepen werden geëvalueerd. Vergelijkingen werden

gemaakt met patiënten die naar CABG werden gerandomiseerd. Complete

revascularisatie werd bereikt in 148 patiënten die naar PCI waren

gerandomiseerd (27%). Bij 147 patiënten (27%) bleef 1 afwijking over,

terwijl in respectievelijk 122 (23%) en 119 (22%) patiënten, 2 en 3 of meer

afwijkingen overbleven. Vijf patiënten (1%) konden niet worden

geclassificeerd. Het percentage sterfte en/of myocardinfarct gedurende 1

jaar was respectievelijk 9,5%, 5,4%, 8,2% en 12,6% in de PCI subgroepen

(P=0,23), en 6,2% in patiënten die werden gerandomiseerd naar CABG

(vergelijking met 3 of meer overgebleven afwijkingen na PCI: p=0,02). ) De

percentages re-interventies gedurende 1 jaar waren 29,7%, 29,3%, 39,3%

and 51,3% (p<0,001), wat in iedere groep hoger was dan na CABG (3,5%;

p<0,001). Complete revascularisatie met PCI bij meervatscoronair lijden

resulteerde niet in een lagere sterfte en/of myocardinfarct in vergelijking

met incomplete revascularisatie. Over het algemeen is de prognose van de

patiënten na PCI vergelijkbaar met CABG. Patiënten met 3 of meer

overgebleven afwijkingen na PCI hebben een slechtere prognose dan

CABG patiënten, door een hogere incidentie van het gecombineerde

eindpunt dood of infarct en meer aanvullende procedures.

In hoofdstuk 4 wordt de 6-maanden klinische uitkomst beschreven bij

patiënten met meervatslijden die werden geincludeerd in de PURSUIT

(Platelet Glycoprotein IIb/IIIa in Unstable Angina: Receptor Suppression

Using Integrelin Therapy) studie. Patiënten werden onderverdeeld in 3


166

groepen volgens de toegepaste behandelstrategie gedurende de eerste 30

dagen na inclusie. De PCI groep werd verder onderverdeeld naar patiënten

die al dan niet met een coronaire stent werden behandeld. De sterfte na 30

dagen was respectievelijk 6,7%, 3,9%, 2,4% en 4,8% voor de

medicamenteuze, PCI (ballon), PCI (stent) en CABG groep (p = 0,002).

Deze verschillen waren nog steeds aanwezig na 6-maanden met

respectievelijk 11,1%, 5,8%, 5,5% en 6,5% sterfte (p = 0,002). Het

optreden van een myocardinfarct binnen 30 dagen was lager bij de

medicamenteuze in vergelijking met de niet medicamenteus behandelde

patiënten, waarbij het hoogste percentage werd gevonden in de CABG

groep (27,7%). Ongeveer de helft van de myocardinfarcten in de PCI en

CABG subgroepen deed zich voor binnen 48 uur na de procedure. Er werd

geconcludeerd dat de geobserveerde verschillen in klinische uitkomsten

mede verklaard kunnen worden door een verschil in basis karakteristieken

en co-morbiditeit.

In hoofdstuk 5, werd de lange termijn overleving bestudeerd in een groep

ongeselecteerde patiënten met coronairlijden die kandidaat waren voor een

invasieve behandeling.

Gegevens van de “DUtch inventory of invasive Coronary Atherosclerosis

Treatments” (DUCAT) studie werden gebruikt. Alle patiënten met significant

coronairlijden die in aanmerking kwamen voor coronaire revascularisatie in

twee tertiaire centra in 1992 werden geincludeerd, vervolg gegevens

werden verzameld in september 2002. In totaal werden 877 patiënten

geincludeerd in deze analyse. De gemiddelde leeftijd was 62 jaar,

chronische stabiele angina pectoris (60%) was de meest gestelde

diagnose. Twaalf procent van de patiënten had diabetes mellitus.


167

Gedurende het vervolgonderzoek bleken 233 patiënten (27%) te zijn

overleden. Voorspellers van lange termijn sterfte waren hogere leeftijd,

diabetes mellitus, perifeer vaatlijden en verminderde linker ventrikel

functie. In vergelijking met medicamenteus behandelde patiënten, hadden

patiënten behandeld met revascularisatie (PCI of CABG) een lagere

lange termijn sterfte (p 0,05). Van de patiënten die werden behandeld met

PCI was 27% overleden, in vergelijking met 24% van de patiënten die een

CABG hadden ondergaan en 36% van degenen die medicamenteus waren

behandeld. Na correctie voor verschillen in basale variabelen bleek

conservatieve therapie echter geen significante voorspeller voor lange

termijn sterfte. Na multivariate analyse bleken hogere leeftijd, verminderde

linker ventrikel functie en diabetes mellitus onafhankelijke voorspellers van

lange termijn sterfte.

De optimale methode van coronaire revascularisatie bij patiënten met

diabetes mellitus blijft controversieel. De doelstelling van hoofdstuk 6 was

om de lange termijn uitkomst te evalueren in de dagelijkse praktijk bij

diabeten met coronairlijden waarbij een invasieve aanpak wordt

overwogen. Hiervoor werden, evenals in hoofdstuk 5, gegevens gebruikt

van de DUCAT studie. Van de geincludeerde patiënten hadden 107

patiënten (12%) diabetes mellitus. Patiënten met diabetes waren ouder en

vaker vrouw. Lange termijn sterfte was hoger in diabeten dan in niet-

diabeten (36% vs. 25%, p=0,01). Deze associatie werd gevonden in

medicamenteus behandelde patiënten (65% vs. 31%, p=0,01) en in

patiënten behandeld met PCI (41% vs. 24%, p=0,02). Er was echter geen

verschil in sterfte tusen patiënten met of zonder diabetes die waren

behandeld met CABG (24% vs. 24%, p=0,89). Multivariate analyse


168

veranderde deze resultaten niet. Er werd geconcludeerd dat diabetes

geassocieerd was met een hogere lange termijn sterfte bij patiënten met

coronairlijden bij wie een invasieve aanpak werd overwogen. Bij patiënten

met diabetes die behandeld worden met CABG lijkt de lange termijn

overleving echter niet slechter te zijn.

In hoofdstuk 7 werden factoren die van belang zijn voor klinische

besluitvorming met betrekking tot revascularisatie bij patiënten met stabiel

coronairlijden onderzocht. De 2936 patiënten met stabiele angina pectoris,

geincludeerd in de Euro Heart Survey on Coronary Revascularisation

(EHS-CR) waren onderwerp van deze analyse. Nadat de diagnose was

bevestigd werd tot een behandeling overgegaan: uitsluitend

medicamenteus of gecominbeerd met revascularisatie middels PCI of

CABG. Teneinde de onafhankelijke relatie te evalueren tussen basale

karakteristieken en een behandelbeslissing (medicamenteus versus PCI,

medicamenteus versus CABG en PCI versus CABG), werd logistische

regressie analyse toegepast. De mediane leeftijd van de patiënten was 64

jaar, 77% was man en 20% van de patiënten had diabetes mellitus.

Medicamenteuze therapie was de voorgenomen therapie in 690 (24%), PCI

in 1503 (51%), en CABG in the overige 743 (25%) patiënten.

Revascularisatie had de voorkeur bij patiënten met ernstiger angina

pectoris, een matig tot groot bedreigd deel van het myocard, en een

behouden linker ventrikel functie en bij patiënten zonder revascularisatie

therapie in de voorgeschiedenis, mits de afwijkingen geschikt waren voor

behandeling. Bij patiënten met meervatslijden of hoofdstam vernauwing en

bij patiënten met bijkomend kleplijden had CABG de voorkeur boven PCI,

mits er voldoende afwijkingen geschikt waren voor CABG. Een PCI had de


169

voorkeur boven re-CABG bij patiënten met een CABG in de

voorgeschiedenis.

Deze analyse van de EHS-CR toont aan dat behandelbeslissingen bij

patiënten met stabiel coronairlijden bepaald worden door meerdere

factoren en globaal in overeenstemming zijn met de richtlijnen. Bij patiënten

met diabetes mellitus echter, bestaat geen voorkeur voor CABG boven

PCI. In de groep van patiënten die werden voorgesteld voor

revascularisatie had PCI de voorkeur bij patiënten met een CABG in de

voorgeschiedenis.

In hoofdstuk 8 werd geëvalueerd of bij patiënten met stabiele angina,

voorkeur voor coronaire revascularisatie middels PCI of CABG beïnvloed

werd door de aanwezigheid van diabetes mellitus en of dit prognostische

implicaties had. De 2928 patiënten met stabiele angina die werden

geincludeerd in de prospectieve Euro Heart Survey on Coronary

Revascularisation werden bestudeerd. Multivariate analyse werd toegepast

teneinde de onafhankelijke relatie tussen diabetes mellitus, behandel

beslissingen en uitkomst na 1 jaar te evalueren. Diabetes mellitus was

gedocumenteerd in 587 patiënten (20%). Deze patiënten hadden meer

uitgebreid coronairlijden. Revascularisatie werd voorgenomen in 74% van

de patiënten met en in 77% van de patiënen zonder diabetes mellitus.

Indien revascularisatie de voorkeur had, was CABG de voorgenomen

behandeling bij 35% van de diabeten en bij 33% van de niet-diabeten.

Multivariate analyses veranderden deze bevindingen niet. In bepaalde

subgroepen werd de behandel beslissing wel beïnvloed door de

aanwezigheid van diabetes mellitus. Bijvoorbeeld, revascularisatie had

vaker de voorkeur bij diabeten met mild hartfalen (91%) in vergelijking met


170

niet-diabeten (67%, p<0,001). Bij patiënten met ernstigere coronair

afwijkingen (hoofdstam-stenose, meervatslijden, of proximale LAD) werden

behandel beslissingen niet beïnvloed door de aan- of afwezigheid van

diabetes mellitus.

Diabetes mellitus was, onafhankelijk van de voorgenomen therapie, niet

geassocieerd met een hogere sterfte, niet-fataal cerebro-vasculair accident

of niet-fataal myocardinfarct na 1 jaar vervolgonderzoek. De incidentie van

het gecombineerde eindpunt was 7,3% in diabeten en 6,8% in niet-

diabeten (na multivariate analyse hazard ratio 1,0 met 95%

betrouwbaarheids interval 0,7 tot 1,4).

Behandel beslissingen betreffende revascularisatie en de keus tussen

CABG of PCI, blijken dus niet beïnvloed door de aan- of afwezigheid van

diabetes mellitus bij patiënten met stabiele angina pectoris en diabetes

mellitus was niet geassocieerd met een slechte prognose.

Slotconclusies en toekomstige ontwikkelingen

Coronaire revascularisatie met behulp van PCI of CABG is een belangrijke

optie voor patiënten met (symptomatisch) coronairlijden. Deze therapie

kan symptomen verminderen en bij sommige patiënten het leven

verlengen. Ofschoon beide revascularisatie therapieën zijn verbeterd

gedurende de laatste decennia, blijft het een uitdaging voor de toekomst

om de meest geschikte therapie op het juiste moment bij de individuele

patiënt toe te passen. Gerandomiseerde studies zijn van groot belang voor

het nemen van beslissingen. Echter, de meeste gerandomiseerde studies


171

hebben ook belangrijke beperkingen, zoals lage externe validiteit en een

beperkte follow-up periode. Verder zijn de meeste gerandomiseerde

studies die betrekking hebben op revascularisatie therapieën niet

geblindeerd en kunnen resultaten worden beïnvloed door “cross-over”.

Daarom moeten deze studies worden gecombineerd met observationele

studies. Aangezien behandelingen en behandelstrategieën veranderen, is

een voortdurende evaluatie van het effect van behandelingen ( van

subgroepen) in de dagelijkse praktijk noodzakelijk.

De uitgebreidheid van het coronairlijden speelt een belangrijke rol in de

keus voor revascularisatie en de selectie voor CABG of PCI. Volledigheid

van revascularisatie kan de prognose verbeteren. Of bij patiënten met

meervatslijden, CABG of PCI de voorkeurstherapie is, dient nog verder te

worden uitgezocht in vele subgroepen van patiënten met coronairlijden, in

het bijzonder bij diabetes mellitus.

Diabetes mellitus is één van de belangrijkste risicofactoren voor

coronairlijden. Daarnaast is er een snelle toename van de prevalentie van

diabetes mellitus in de algemene populatie. Het is nog steeds onzeker of

PCI dan wel CABG de voorkeursbehandeling is bij de meerderheid van

patiënten met diabetes mellitus en coronairlijden. Vermoedelijk zullen anti-

restenotische, drug-eluting stents (DES) de resultaten van PCI in

vergelijking met CABG bij deze patiënten verbeteren. Toekomstige studies

als de CARDia, FREEDOM en BARI 2D, waarin patiënten met

coronairlijden en diabetes mellitus worden geincludeerd, en vervolgens

gerandomiseerd naar CABG of PCI met DES en moderne anti-plaatjes

therapie, inclusief GP IIb/IIIa inhibitors en clopidogrel, zullen waarschijnlijk

meer duidelijkheid verschaffen in dit PCI-versus-CABG debat.

Curriculum Vitae ___________________________________________________________________

173

Curriculum Vitae

Arno Breeman was born on October 18th 1960 in Zwijndrecht, The

Netherlands. After following secondary school at the Melanchton

Scholengemeenschap in Rotterdam he studied medicine at the Erasmus

University Rotterdam and graduated in 1987. After graduation he fulfilled

his military service at the military medical service in Breda. Subsequently

he worked as house-officer cardiology at the Merwede Hospital in

Dordrecht and started his first research activity on scintigrafic imaging

techniques under the directon of dr. M.G. Niemeyer and Prof. dr. E.E. van

der Wall. In 1990 he started as research fellow at the Thoraxcentre

Rotterdam under the direction of Prof. dr. P.W. Serruys and was involved in

the Coronary Artery Bypass Revascularisation Investigation (CABRI) trial

and the Arterial Revascularization Therapies Study (ARTS). At the same

time he took part in the DUtch Inventory of Coronary Atherosclerosis

Treatments (DUCAT) steering committee under direction of Prof. dr. H.

Rigter. In 1993 he fulfilled his residency in internal medicine at the

Merwede Hospital in Dordrecht under direction of dr. B.A. de Planque as

part of his training as cardiologist at the Thoraxcentre Rotterdam under the

direction of Prof. dr. J.R.T.C. Roelandt and graduated in 1998. In 1999 he

became staff member cardiology in Zwolle. Here, he completed the 10-year

follow-up of the DUCAT study and was involved in the Euro Heart Survey

(EHS), which resulted in studies presented in this thesis. On the moment

he develops a digital echocardiography system and is involved in the daily

practice of cardio-pulmonal surgical care. Arno is married with Loes van

Curriculum Vitae ___________________________________________________________________

174

Ruitenburg. They have three children, Evert (1992), Karel (1995) and Floris

(1998). He likes to listen to a broad range of music, and tries to play music

himself. Furthermore, he spends his time with his family, Loes and the

boys, particularly in outdoor activities.

List of Publications ___________________________________________________________________

175

List of Publications

Neijens HJ, Breeman A, Sinaasappel M, Degenhart HJ, Kerrebijn KF, Visser HKA. The Nutritional status and the pulmonary abnormalities as a function of age in cystic fibrosis and healthy children. Lawson D, ed. Cystic Fibrosis: Horizons. New York: Wiley, 1984:85.

Niemeyer MG, Zwinderman AH, Cramer MJ, van der Wall EE, Verzijlbergen FJ, Breeman A, Ascoop CA, Pauwels EKJ. Improvement of diagnosis in the non-invasive assessment of coronary artery disease: enhanced evaluation of quantitative exercise Thallium-201 imaging by multivariate analysis. Cardiovasc Res 1990;10:804-812.

Breeman A. Dipyridamole stress testing for the diagnosis of coronary artery disease (congress report). Ned Tijdschr Cardiol 1991;2:70-72.

Breeman A. First Biennial symposium of the European Working Group on Nuclear Cardiology (congress report). Ned Tijdschr Cardiol 1991;4:170-172.

Oosterhuis WP, Zwinderman AH, Kuyper A, Breeman A, Niemeyer MG. Prognostic value of Thallium 201 myocardial scintigraphy. Circulation 1991;84:2203.

Serruys PW, Breeman A. Coronary angioplasty, clinical application 1991. Neth J Cardiol 1991;4:146-154.

Breeman A, Serruys PW. DUAST, een gemiste kans. Ned Tijdschr Cardiol 1991;7:251-253.

Serruys PW, Breeman A. Coronary angioplasty, clinical application 1991. Cardioscopies 1991;20:10-20.

Breeman A, Niemyer MG, Fintelman H, Ernst JMPG, Ascoop CAPL. De betekenis van ST-elevatie bij inspannings-electrocardiografie. Ned Tijdschr Cardiol 1992;6:175-179.

Serruys PW, Breeman A. Coronary angioplasty, clinical application 1991, longterm follow-up results and detection of restenosis: guidelines for aviation cardiology. An European view. Eur Heart J 1992;13:76-88.


176

Rigter H, Meijler AP, Breeman A, Sprenger ML, Bernstein SJ. DUCAT: an observational study of clinical decision making in patients with angiographically established coronary artery disease. Neth J of Cardiol 1992;5:131-136.

Breeman A, Serruys PW, van den Brand MJBM, van Herwerden LA, Strikwerda S, K. Meeter, J.M.P.G. Ernst, J.J.R.M. Bonnier, J.R.T.C. Roelandt. Onderzoek naar de klinische waarde van invasieve behandelingen aan coronaire arterien. Een overzicht (I). Ned Tijdschr Cardiol 1992;6:199-208.

Breeman A, Serruys PW, van den Brand MJBM, van Herwerden LA, Strikwerda S, Deckers J, Bonnier JJRM, Ernst JMPG, Roelandt JRTC. Angioplastiek versus coronairchirurgie bij meervatscoronairlijden. Een overzicht (II). Ned Tijdschr Cardiol 1992;8:238-246.

Oosterhuis WP, Niemeyer MG, Kuijper AFM, Zwinderman AH, Breeman A, Ascoop CAPL, Verzijlbergen FJ, van der Wall E, Pauwels EKJ. Evaluation of the incremental diagnostic value and impact on patient treatment of scintigraphy. J of Nucl Med 1992;33:1727-1734.

Oosterhuis WP, Breeman A, Niemeyer MG, Zwinderman AH, Kuyper A, van der Wall EE, Tijssen JGP, Pauwels EKJ. De prognostische waarde van een normaal thallium-201-inspanningsscintigram: nadelig invloed van B-blokkergebruik op testresultaat. Ned Tijdschr Geneeskd 1992;136:714-720.

Oosterhuis WP, Niemeyer MG, Breeman A, Zwinderman AH, Ascoop CAPL, Verzijlbergen FJ, Kuijper AFM, van der Wall EE, Pauwels EKJ. Exercise-201 - thallium scintigrafy; Evaluation of the additional diagnostic value. Nucl Med Commun 1993;14:87-95.

Oosterhuis WP, Breeman A, Niemeyer MG, Zwinderman AH, Kuyper A, van der Wall EE, Tijssen JPG, Pauwels EKJ. Patients with a normal thallium-201 scintigram: Always a good prognosis? Eur J of Nucl Med 1993;20:151-158.

Breeman A, Serruys PW. Indications for heart catherisation after PTCA or CABG. Int J Cardiac Imag 1993;71:71-76.

Breeman A, Serruys PW, van den Brand MJBM, Deckers JW, van Herwerden LA, Roelandt JRTC. Complicaties kort na percutane transluminale angioplastiek en coronairchirurgie bij 183 vergelijkbare patiënten met meervatscoronairlijden. Ned Tijdschr Geneeskd 1994; 21:1074-1079.

Van Berkel TFM, Erdman RAM, Breeman A, Boersma H, van den Brand MJBM. Kwaliteit van leven na coronairchirurgie en na ballonangioplastiek; meer pijn op de


177

borst en sociale geremdheid na angioplastiek. Ned Tijdschr Geneeskd 1995;34:1733-1736.

Rigter H, Meijler AP, Scholma J, McDonnell J, Breeman A. DUCAT (DUtch Inventory of Invasive Coronary Atherosclerosis Treatment). Passende indicaties voor behandeling van patiënten met coronairlijden. Erasmus Universiteit Rotterdam, Instituut Maatschappelijke Gezondheidszorg/Institute voor Medische Technologie Assessment. 1995 ISBN 90-72245-63-6.

Breeman A, Scholma JK, Meijer AP, Rigter H. Patiënten die in de hartcentra worden besproken het DUCAT onderzoek. Hartbulletin 1996;6:187-190.

Meijler AP, Rigter H, Bernstein SJ, Scholma JK, McDonnell J, Breeman A, Kosecoff JB, Brook RH. The appropriateness of intention to treat decisions for invasive therapy in coronary artery disease in the Netherlands. Heart 1997;77:219-224.

Serruys PWS, Unger F, van Hout B, van der Brand M, van Herwerden L, van Es GA, Bonnier JJRM, Colombo A, Morice MC, Simon R, Wijns W, Kremer J, Mohr F, Petterson G, Santoli C, Breeman A, Vandormael M, Marshal P, Madona 0, Firth B, Hugenholtz PG. The ARTS (Arterial Revascularisation Therapies Study (ARTS). Background, goals and methods. Int J Cardiovasc Interventions 1999;2:41-50.

Serruys PW, Unger F, van Hout BA, van den Brand MJ, van Herwerden LA, van Es GA, Bonnier JJ, Simon R, Cremer J, Colombo A, Santoli C, Vandormael M, Marshall PR, Madonna O, Firth BG, Breeman A, Morel MA, Hugenholtz PG. The ARTS Study (Arterial Revascularisation Therapies Study). Semin Interv Cardiol 1991;4:209-19

Breeman A, Boersma E, Brand van den MJ, Herwerden van L, Serruys PW, for the CABRI-investigators. Completeness of revascularisation by percutaneous coronary intervention. Neth Heart J 2001;9:3-9.

Domburg RT van, Foley DP, Breeman A, van Herwerden LA, Serruys PW. Coronary artery bypass graft surgery and percutaneous transluminal coronary angioplasty. Twenty- year clinical outcome. Eur Heart J 2002;23:543-9.

Breeman A, Mercado N, Lenzen M, van den Brand MM, Harrington RA, Califf RM, Topol EJ, Simoons ML, Boersma E; PURSUIT Investigators. Characteristics, treatment and outcome of patients with non-ST-elevation acute coronary syndromes and multivessel coronary artery disease: observations from PURSUIT (platelet glycoprotein IIb/IIIa in unstable angina: receptor suppression using integrelin therapy). Cardiology 2002;98:195-201.


178

Brand van den MJ, Rensing BJWM, Morel MM, Foley DP, Valk de V, Breeman A et all. The effect of the completeness of revascularisation on event free survival at 1 year Follow-up in the ARTS trial. J Am Coll Cardiol 2002; 39: 559-64.

Breeman A, Ottervanger JP, Boersma E, De Luca G. Coronary revascularisation for Non-ST Elevation Acute Coronary Syndrome. State of the Art. J Cardiovasc Med: Accepted nov 2005.

Breeman A, Timmer J, Ottervanger JP, Kolkman E, Kluiver de E, Rigter H, Boonstra P, Zijlstra F. Long-term follow-up after invasive approach of coronary artery disease in daily practice. Int J Cardiol 2005;105:186-91.

Timmer JR, Breeman A, Ottervanger JP, Kluiver de E, Boonstra P, Zijlstra F. Long-term survival after coronary revascularisation in patients with diabetes mellitus. Submitted.

Breeman A, Hordijk M, Lenzen M, Hoeks S, Ottervanger JP, Bertrand M, Sechtem U, Zaliunas R, Legrand V, Boer de MJ, Stahle E, Mercado N, Wijns W, Boersma E. Treatment decisions in patients with stable coronary artery disease in a broad range of European practices. Insights from the Euro Heart Survey on Coronary Revascularisation. J Thor Cardiov Surg (in press).

Breeman A, Bertrand ME, Ottervanger JP, Hoeks S, Lenzen M, Sechtem U, Legrand V, Boer de MJ, Wijns W, Boersma E. Diabetes does not influence treatment decisions regarding revascularisation in patients with stable coronary artery disease. On behalf of the investigators of the Euro Heart Survey on Coronary Revascularisation. Submitted.

Dankwoord___________________________________________________________________

179

Dankwoord

Het tot stand komen van dit proefschrift is teamwerk en het resultaat van de

inspanningen van velen. Graag wil ik iedereen bedanken die op welke

manier dan ook hieraan heeft bijgedragen. Een aantal personen wil ik in het

bijzonder noemen.

In de eerste plaats wil ik mijn beide copromotoren danken. Het afronden

van dit proefschrift bleek een “pièce de résistance” te zijn. Dankzij jullie

inzet en steun werd het mogelijk deze weerstand te overwinnen, waarvoor

mijn grote dank.

Dr. J.P. Ottervanger, beste Jan Paul. Collega, maat, vriend en nu ook

copromotor. Je kwam, zag en overwon de obstakels van een proefschrift

dat maar niet verder kwam. Opmerkelijk is je gave vele denkprocessen

tegelijkertijd helder uit te kunt uitvoeren, een eigenschap waarvan velen

profijt hebben. Ik waardeer ook zeer onze gesprekken in de (vroege)

ochtenduren waarin we ook tijd hadden om over andere zaken dan het

proefschrift te praten. Bedankt voor de vele goede dingen die je voor me

hebt gedaan, ik verheug me er op nog lang met je te mogen samenwerken.

Dr. E. Boersma, beste Eric. We kennen elkaar al zo’n 15 jaar, het begin

van onze carrière in het Thoraxcentrum. In deze periode hebben we heel

wat analyses uitgevoerd en besproken. Nog steeds kan ik je redenaties

een heel stuk volgen maar moet ik regelmatig afhaken wanneer ik probeer

jouw wereld van wiskundige en statistische modellen te betreden. Het is

altijd een groot genoegen met je samen te werken en ik ben je zeer

Dankwoord___________________________________________________________________

180

dankbaar voor je onvoorwaardelijke inzet en trouw in deze lange

samenwerking die uiteindelijk mede heeft geleid tot de bekroning in de

vorm van dit proefschrift.

Prof. dr. F. Zijlsta, beste Felix. Met bewondering voor je (wetenschap-

pelijke) werk ben ik destijds naar Zwolle gekomen. Je vertrek naar

Groningen betreurde ik, omdat we je in Zwolle moesten missen. Het was

echter een logisch vervolg op hetgeen je deed en doet en is je uiteraard

van harte gegund. Ik ben blij dat jij mijn promotor kan en wil zijn. Ik dank je

ook voor al je adviezen in het voortraject en hoop dat we in de toekomst

ook nog kunnen samenwerken.

De leden van de beoordelingscommissie, prof. dr. D.J. van Veldhuisen,

prof. dr. P.W. Boonstra en prof. dr. M.L. Simoons dank ik voor hun

bereidwilligheid dit proefschrift kritisch te beoordelen. Beste Maarten, ik ben

blij dat we elkaar via dit proefschrift weer hebben teruggevonden.

De beide paranimfen wil ik reeds bij voorbaat danken voor het werk dat ze

nog moeten doen. Drs. P.P.H.M. Delnoy, beste Peter Paul, maatje van het

eerste uur in Zwolle. Ik ben blij dat je me op de promotie terzijde wil staan,

en voel me daar in meer dan één opzicht door gesteund. Je collegialiteit en

vriendschap zowel in als buiten het ziekenhuis zijn voor mij van grote

waarde. Dr. E.P. de Kluiver, beste Ed, als co-assistent in Goes leerde ik je

reeds kennen. Nog steeds ben ik je voor veel dingen dankbaar. Je bent

een stabiliserende factor binnen onze maatschap, soms zie ik je bijna als

coach, waarbij je het goede in een ieder naar boven weet te halen. Je

kennis en keuze van wijnen maakt daarbij iedere maatschapvergadering

Dankwoord___________________________________________________________________

181

weer tot een lust. Je stimuleerde me verder te gaan met onderzoek op een

moment waarop ik daar zelf twijfels over had.

De DUCAT studie werd destijds opgezet door prof. dr. H. Rigter. Beste

Henk, het was een voorrecht dat ik “in de keuken” van een groot

psycholoog, diplomaat en besliskundige mocht kijken. Ik wil je bedanken

voor het vertrouwen dat je in me hebt gesteld en het feit dat ik de gegevens

mocht gebruiken voor een follow-up studie, die overigens niet mogelijk was

geweest zonder het Zwolse onderzoeksinstituut “Diagram” en het intensief

naspeuren door Tini Eenhuizen, waarvoor dank. De statistische analyses

werden voor een groot deel “op Diagram” gedaan door Evelien Kolkman,

vaak bijna onopvallend een goede hulp. Bij de analyses en het opschrijven

van de resultaten van patiënten met diabetes ben ik dank verschuldigd aan

Jorik Timmer. Beste Jorik, ik hoop dat je weer spoedig de weg naar Zwolle

weet terug te vinden!

Uiteraard ben ik dankbaar dat ik gegevens van de Euro Heart Survey

mocht gebruiken. Special thanks to my European friends William Wijns en

Michel Bertrand for supporting these studies! Mattie Lenzen, Sanne Hoeks,

Metske van der Wal, Henriette Tebbe, Marie Angèle Morel, Eline

Montauban van Swijndregt, Ron van Domburg en medewerkers klinische

epidemiologie van het Erasmus MC, dank voor jullie steun en

inspanningen.

Aan mijn opleiding tot cardioloog en mijn wetenschappelijke vorming

hebben tal van mensen bijgedragen. Zonder volledig te zijn en in

willekeurige volgorde wil ik met name noemen Menco Niemeyer, Ernst van

Dankwoord___________________________________________________________________

182

der Wall, Patrick Serruys, Jos Roelandt, Maarten Simoons, Marcel van den

Brand, Jaap Deckers, Lex van Herwerden, Annejet Meijler, Jan Scholma,

Walter Hermans en vele, vele anderen waarvan ik heb mogen leren.

Al mijn maten cardiologie in Zwolle dank ik voor hun vertrouwen en de tijd

die ze me gaven om aan het proefschrift te werken. Ik hoop uiteraard dat

met name de interventiecardiologen, Jan Hoorntje, Menko Jan de Boer,

Harry Suryapranata, Arnoud van ‘t Hof, Jan-Henk Dambrink, Marcel

Gosselink en Jan Paul Ottervanger iets zullen hebben aan de studies die in

het proefschrift staan beschreven. Vooral Harry wens ik sterkte toe met de

realisatie dit jaar van “ons ziekenhuis” in Djakarta, “keberhasilan!”. De

elektrofysiologen, Henk Oude Luttikhuis, Anand Ramdat Misier, Willem

Beukema, Peter Paul Delnoy en Arif Elvan wens ik sterkte in deze tijden

waarin de ritmecardiologie in een stroomversnelling lijkt te zijn gekomen. Ik

hoop dat Willem en Peter Paul de tijd vinden en krijgen om hun eigen

proefschrift af te ronden. Willem Jan Louridtz, je bent altijd bereid “klusjes

op te knappen”, of het nu een consult of een “slokdarm-echo” of een

catheterisatie is. Dank daarvoor, we zullen je missen als je volgend jaar

gaat pensioneren. Angela Maas, mijn enige vrouwelijke maat, wens ik

sterkte toe met de laatste fase van het schrijven van haar eigen

proefschrift. Ik waardeer de manier waarop jij je inzet voor het verbeteren

van het welzijn van de vrouw met hart- en vaatziekten en denk dat je je

verdediging straks met glans zal doorstaan. Cardiologie in Zwolle zou niet

kunnen bestaan zonder steun van onze maat en maatschapcoördinator Ed

de Kluiver.

Dankwoord___________________________________________________________________

183

Uiteraard zijn er vele personen in het Zwolse ziekenhuis die mij direkt of

indirekt geholpen hebben in de achterliggende jaren: thoraxchirurgen,

anesthesiologen, arts-assistenten etc, etc. Ik kan ze niet allen bij naam

noemen, maar ieder van harte bedankt. Dat geldt ook voor het

management, vooral onze zorgmanager Fenna Eefting die ik wel bij naam

wil noemen. Beste Fenna, de aangeboren argwaan van medisch

specialisten tegen managers hebben wij bij jouw inmiddels wel verloren!

Een speciaal woord van dank voor Vera Derks. Vera, jij bewaakt alle

wetenschappelijke output van Zwolle, ook voor mij was je onmisbaar bij het

aanbieden van artikelen, de voorbereidingen van het proefschrift en het

aanvragen van subsidies. Thea Schenk bedankt voor de “finishing touch”.

Jouw ervaring bij Elsevier bleek van grote waarde voor de spelling en lay-

out van het proefschrift. Mark Retera wil ik danken voor het vervaardigen

van de cartoon op de cover van dit proefschrift, om je “Dirk-Jan” strip wordt

bij ons thuis gevochten!

De ontwikkeling en uitwerking van dit proefschrift heeft zich voor een groot

gedeelte in mijn “vrije tijd” afgespeeld. Familie, vrienden en kennissen

hebben dit proces van nabij kunnen meemaken. De vriendschap en

harmonie die ik bij jullie vond gaven me de balans om dit proefschrift af te

ronden. Een paar vrienden van “het eerste uur” wil ik bij name noemen.

Charles, wijze vriend, jij begreep dat ik rust nodig had om een doorstart van

het proefschrift te maken en gaf dat in jullie huisje in Zeeland. Beste Frank,

dank voor je langdurende vriendschap en je vermogen om altijd maar weer

te blijven relativeren. Beste Gijs, jij liet me overtuigend zien dat geloof in je

zelf tot grote dingen kan leiden. Je bent een goed voorbeeld voor me

geweest. Beste Donald, voor jou kwamen de nieuwste cardiologische

Dankwoord___________________________________________________________________

184

ontwikkelingen helaas te laat. Jouw vriendschap is eeuwig. Ik zal je missen

op mijn feest.

Aan al mijn andere vrienden, waaronder onze “huis-filosofen” Jacques en

Koert en de unieke “Berkumse Boys”, dank voor jullie interesse en

medeleven, ik ben weer van de partij!

Speciale dank aan mijn ouders. Ik heb van jullie alle liefde en

mogelijkheden gekregen om mijn talenten te kunnen benutten. Ik ben jullie

daar zeer dankbaar voor. Ik ben dan ook heel blij dat jullie de bekroning

van mijn wetenschappelijk werk kunnen meemaken. Lieve zussen, je

broertje heeft het gefixed, dank voor jullie steun. Beste neven, bedankt voor

jullie nimmer aflatende snelle hulp bij “spoedeisende” digitale problemen.

Lieve Loes, je bent een fantastische vrouw. Ik dank je met heel mijn hart

voor je nimmer aflatende vertrouwen en steun, zonder jouw inzet was dit

proefschrift nooit tot stand gekomen. Lieve Evert, Karel en Floris,

vertederend hoe jullie met je vader ieder op zijn eigen manier hebben

meegeleefd. Wat kunnen jullie goed stil zijn als het nodig is. Gelukkig is dat

nu voorbij. We gaan weer leuke dingen doen en voortaan blijft de laptop

tijdens vakanties thuis!!

Documents

University of Groningen Invasive treatment of …Invasive treatment of coronary artery disease: lessons from clinical trials and clinical practice Proefschrift ter verkrijging van