University of Groningen

Individual approach towards optimal oral anticoagulation Veeger, Nicolaas Johannes Gerardus Maria

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below. Document Version Publisher's PDF, also known as Version of record

Publication date: 2010

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA): Veeger, N. J. G. M. (2010). Individual approach towards optimal oral anticoagulation. s.n.

Copyright Other 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 the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

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

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

Download date: 23-09-2021 Chapter 6

Chapter 6

Vitamin K Antagonists or Aspirin Combined with Dipyridamole Not Superior to Aspirin Alone; Fourteen Year Follow-up Results of CABADAS

Nic J.G.M. Veeger Felix Zijlstra Hans L. Hillege Jan van der Meer

Submitted

- 93 - CABADAS 14-year follow-up study

SUMMARY

Background. Secondary prophylaxis using aspirin is standard of care after coronary- artery-bypass-graft surgery. Limited data are available for long-term results. We evaluated the effect of aspirin, aspirin with dipyridamole and vitamin K antagonists (VKA) on 14-year clinical outcome of patients included in the CABADAS study. Methods. All 726 Dutch patients in whom antithrombotic therapy with aspirin (N=248), aspirin with dipyridamole (N=234) or VKA (N=244) was randomly allocated, were included. Primary endpoint was occurrence of Major Adverse Cardiac Events (MACE). Outcome was retrospectively evaluated during 14 year follow-up. Results. Cumulative incidences for MACE over 14 years were 49%, 50% and 59% in patients treated with aspirin, aspirin with dipyridamole and VKA, respectively. Although overall occurrence of MACE did not significantly differ between the three treatment groups (P=0.12), patients treated with VKA were at higher risk of MACE than patients treated with aspirin with dipyridamole (HR=1.3, 95%CI, 1.0-1.8; P=0.041) and patients treated with aspirin alone (HR=1.1, 95%CI, 0.86-1.5; P=0.37). This difference was attributed to an increased risk of repeat revascularization in patients treated with VKA, without any differences in cardiac death and myocardial infarction between the three treatment groups. However, the observed high rate of repeat revascularization in patients treated with VKA could reflect an a priori increased probability for repeat revascularization due to the specific conditions surrounding VKA therapy (i.e. more intense patient-doctor contacts). Conclusions. This study with 14-year clinical outcome provide further evidence for the use of aspirin as secondary prophylaxis after coronary-artery-bypass-graft surgery.

- 94 - Chapter 6

INTRODUCTION

Several prospective controlled clinical trials have evaluated efficacy and safety of antithrombotic drug therapies for secondary prophylaxis after coronary-artery-bypass- graft surgery to prevent vein graft closure [1]. This resulted in a high level of evidence, identifying aspirin as standard of care [2,3]. To maximize the effect of anticoagulant treatment it must be initiated already early after surgery [4-6]. In this respect, current guideline stipulates the initiation of aspirin within 48 hours after surgery and to be continued indefinitely [4,5]. When started more than 48 hours after surgery, efficacy of aspirin is hampered and early graft occlusion more frequently observed [7]. One of the clinical trials was CABADAS (CABADAS = prevention of Coronary Artery Bypass graft occlusion by Aspirin, Dipyridamole and Acenocoumarol/phenprocoumon Study). CABADAS evaluated the effect of aspirin, aspirin with dipyridamole and vitamin K antagonists (VKA) on one-year graft patency and early clinical outcome. The CABADAS study concluded that there was no convincing evidence that the addition of dipyridamole to a low dose of aspirin improves one-year vein-graft patency after coronary artery bypass grafting. Oral anticoagulants, compared with aspirin, provided no benefit [8]. In many of the clinical trials that contributed to the evidence of aspirin as primary choice for secondary prophylaxis after coronary-artery-bypass-graft surgery, data on long-term clinical outcome are limited. As in CABADAS, mainly early effects on graft patency and short- and mid-term clinical outcome are available. Therefore, in the present study we evaluated the clinical outcome of patients included in the CABADAS study after an extended follow-up up to 14 years. These patients were originally treated with aspirin, aspirin with dipyridamole or VKA.

PATIENTS AND METHODS

The Extended CABADAS study is a retrospective follow-up study in 726 Dutch participants of the CABADAS study (CABADAS = prevention of Coronary Artery Bypass graft occlusion by Aspirin, Dipyridamole and Acenocoumarol/ phenprocoumon Study). A priori, it was decided to only include the Dutch participants for the extension of the CABADAS study, for logistical reasons. The original CABADAS study was an international multicenter randomized controlled

- 95 - CABADAS 14-year follow-up study

clinical trial evaluating the effects of low dose aspirin, aspirin combined with dipyridamole and vitamin K antagonists on graft patency and clinical outcome at one year. In this study, 912 patients were evaluated after one year of study treatment. The design of the study is described in great detail elsewhere [8]. Initial study protocol was approved by the Ethics Committee of each participating center and all patients gave written informed consent.

Patients All patients underwent elective CABG for disabling angina between July 1987 and August 1990. Surgery was performed using routine techniques in each participating center. They all received saphenous vein grafts, and in addition, internal mammary artery (IMA) grafts could be used at the discretion of the surgeon. Heparin, administrated during surgery was antagonized by protamine sulphate at the end of the procedure, unless it was continued because of the introduction of an intra-aortic balloon pump. Main exclusion criteria were age over 70 years, previous or concurrent cardiac surgery, unstable angina at the time of surgery (< 2 days) or myocardial infarction within 7 days prior to surgery and severe concomitant disease.

Randomization Study treatment was initiated after random allocation, i.e. aspirin, aspirin with dipyridamole or vitamin K antagonists (acenocoumarol or phenprocoumon) to all eligible patients, prior to surgery. To ensure a balanced treatment in all participating hospitals, allocation was stratified by study site.

Follow up After discharge from the hospital, patients were seen for the duration of one year at a 3-month interval. The extended long-term follow-up was retrospectively performed for all patients up to June 2002, through telephonic interviews of the patients using a standard questionnaire and by reviewing the medical files of their cardiologists and/or general practitioners. This evaluation was performed for the first time 2 to 3 years after surgery and repeated 12 years thereafter.

- 96 - Chapter 6

Clinical endpoints The primary endpoint of the extended follow-up study was the composite of Major Adverse Cardiac Events (MACE), i.e. cardiac mortality, myocardial infarction and coronary repeat revascularization. Cardiac mortality was defined as mortality due to acute myocardial infarction, acute cardiac arrest, sudden death and mortality caused by progressive congestive heart failure. In case of an unknown cause of death, it was classified as cardiac death. For myocardial infarction, pathological Q-waves had to be present on the ECG in combination with CPK and CPK-MB serum levels that exceeded the upper limits of normal ranges. Repeat revascularization was defined as percutaneous coronary intervention (PCI) or redo-CABG during follow-up. From the original CABADAS study, patients’ characteristics, general and disease specific medical history and clinical outcome events were available up to one year after discharge. Adjudication of these events was performed by an blinded classification committee, without knowledge of the patient’s treatment assignment.

Anticoagulation data Patients allocated to VKA therapy were referred to specialized Thrombosis Services for frequent control of the level of anticoagulation. The VKA of choice was acenocoumarol or phenprocoumon. The target range used at the time of the original study was INR 2.8 to 4.8.

Statistical analysis The primary objective of the study was to evaluate the effect of three anti-thrombotic drugs on long-term clinical outcome in patients who underwent CABG for disabling angina. Differences between groups were evaluated by survival analysis using an Intention-To-Treat approach. In this, patients were evaluated by their originally allocated treatment. The event-free survival was graphically depicted using the method of Kaplan-Meier. A multivariable Cox proportional hazards regression analysis was performed to estimate adjusted hazard ratios (HR). Predefined risk factors for MACE were age, sex, Body Mass Index, smoking, diabetes mellitus, hypertension, hypercholesterolemia, history of arterial thrombosis, NYHA classification, concomitant cardiovascular medication, severity of vessel disease, type of grafts used (vein or arterial) and number of grafts used.

- 97 - CABADAS 14-year follow-up study

Secondary survival analyses were performed on separate components of MACE using the same strategy as in the primary analysis, as well as a sensitivity analysis based on an “As Treated” approach.

Table 1 Baseline characteristics of patients and anticoagulation data.

Aspirin Aspirin + VKA† Total P- Dipyridamole value N=248 N=234 N=244 N=726

Patients’ characteristics Age, years * 59 ± 7.7 59 ± 7.7 58 ± 7.5 58 ± 7.6 0.46 Male gender, % 86 82 88 86 0.21 Weight, kg * 77 ± 10.4 77 ± 10.2 78 ± 9.9 78 ± 10.1 0.93 Height, cm * 173 ± 7.6 173 ± 8.1 173 ± 7.8 173 ± 7.9 0.99 Angina NYHA class, % II 29 26 25 27 0.92 III 55 56 57 56 IV 16 18 18 17 Myocardial infarction, % 52 48 51 51 0.62 CVA or TIA, % 4 3 1 3 0.16 Hypertension, % 29 30 37 32 0.11 Diabetes mellitus, % 8 9 9 9 0.93 Hyperlipidemia, % 35 34 43 37 0.12 Smoking (current/past), % 36/50 34/46 40/48 37/48 0.15 Concomitant drugs at admission, % 80 76 77 78 0.46 beta blockers 66 62 64 64 0.75 calcium antagonists 81 80 83 81 0.61 nitrates 15 13 17 15 0.53 diuretics 10 10 11 10 0.95 lipid lowering drugs 4 4 5 4 0.99 ACE inhibitors 2 2 2 2 0.99 digitalis

Surgical data Vessel disease, % single 3 1 2 2 0.23 double 39 39 33 37 triple 58 60 65 61 Number of grafts * 2.2 ± 0.9 2.5 ± 1.2 2.3± 0.7 2.3 ± 0.9 0.003 Graft type, % vein grafts only 29 29 42 40 0.70 vein plus arterial grafts 61 61 58 60 Concomitant drugs at discharge, % beta blockers 36 36 35 36 0.97 calcium antagonists 8 10 14 11 0.11 nitrates 1 1 1 1 0.83 diuretics 15 16 14 15 0.76 lipid lowering drugs 5 3 7 5 0.091 ACE inhibitor 2 3 5 3 0.28 digitalis 13 15 12 13 0.54 * mean ± SD; † Vitamin K antagonists.

- 98 - Chapter 6

All analyses performed to test the (null-) hypothesis of no differences between groups, were two-sided. A p-value < 0.05 was considered statistically significant. For all analyses, commercially available computer software (Statistical Analysis System version 9.1, SAS Institute, Cary, North Carolina) was used.

RESULTS

Patients’ characteristics The study cohort comprised of 726 patients from the 912 eligible patients in the original CABADAS study. These were the Dutch patients, of whom 248 were allocated to receive aspirin plus placebo, 234 aspirin with dipyridamole and 244 vitamin K antagonists. Baseline characteristics of the patients are summarized in Table 1. Eighty-eight % of the patients were males and the mean age at the time of surgery was 59 years. All patients had disabling angina, with 56 % NYHA class III and 17% NYHA class IV. Fifty-one % had a previous myocardial infarction, hypertension was diagnosed in 32% and hyperlipidemia in 37%. Bypass grafting was performed on all three coronary arteries in 61%, in 37% on two of the three arteries and a single artery in 2% of patients. Arterial grafts, together with vein grafts, were used in 60% of patients. Due to the randomized study design with a balanced treatment allocation over participating study sites, the three treatment groups were highly comparable at baseline.

Clinical outcome MACE Median duration of the extended follow-up was 12.6 years (range 0.08 -14.8) for the aspirin group, 12.7 years (range 0.02 - 14.8) for the aspirin with dipyridamole group and 12.4 years (range 0.04 - 15.1) for the VKA group. The cumulative incidence of MACE over the 14-year period was 49% in patients treated with aspirin, 50% in patients treated with aspirin with dipyridamole and 59% in patients treated with VKA. These observed differences between aspirin, aspirin with dipyridamol and VKA were not significant (P=0.12; from multivariable Cox regression), although the patients treated with VKA were at higher risk for MACE than the patients treated with aspirin with dipyridamole and aspirin (HR=1.3, 95%CI, 1.01 to 1.8; P=0.041 and HR=1.1, 95%CI, 0.86 to 1.5; P=0.37, respectively;

- 99 - CABADAS 14-year follow-up study

aspirin compared to aspirin with dipyridamole HR=1.2, 95%CI, 0.89 to 1.6; P=0.24; see Figure 1).

60% Aspirin Aspirin + dipyridamole 50% VKA

40%

30%

MACE

20%

10% P = 0.12 0% 0 2 4 6 8 101214

Patients at risk: Time (years) 248 208 192 181 158 139 115 15 234 205 188 171 157 142 125 19 244 200 183 162 156 134 116 16

Figure 1 Kaplan Meier of the composite of cardiac death, myocardial infarction or repeat revascularization (MACE) after coronary-artery-bypass-graft surgery.

As presented in Figure 2, significantly associated with MACE were Age at time of surgery (HR 56 to 62 years = 1.2, 95%CI, 0.90 to 1.6, P=0.22 and HR > 62 years = 1.4, 95%CI, 1.1 to 1.9; P=0.009, compared to patients up to the age of 55 years), NYHA class III or IV (HR=1.4, 95%CI, 1.1 to 1.8; P=0.018), Use of digitalis (HR=2.3, 95%CI, 1.3 to 4.3; P=0.006), Beta-blocking agents (HR=0.73, 95%CI, 0.56 to 0.94; P=0.014), and Lipid lowering drugs (HR=0.55, 95%CI, 0.35 to 0.87; P=0.011), all prior to surgery and Type of graft used during surgery (HR vein graft only = 1.3, 95%CI, 1.1 to 1.6; P=0.022). The number of grafts used during surgery were not associated with MACE (HR 1 or 2 grafts vs. 3 or more = 1.0, 95%CI, 0.8 to1.3; p=0.75).

Components of MACE Regarding the components of MACE separately, cardiac death had occurred in 135 patients with a cumulative incidence over the 14-year period of 23%. In the patients

- 100 - Chapter 6

treated with aspirin, aspirin with dipyridamole and VKA, this was 21%, 24% and 25%, respectively (HR=0.96 (95%CI, 0.64 to 1.5; P=0.85) and HR= 1.0 (95%CI, 0.67 to 1.5; P=0.94) of aspirin and aspirin with dipyridamole, compared to VKA, P=0.96 (aspirin compared to aspirin with dipyridamole HR=0.95, 95%CI, 0.62 to 1.4; P=0.79); see Figure 3).

Variable Univariate P Univariate Hazard Ratio Adjusted P Hazard Ratio Hazard Ratio (95%CI) (95%CI)

Allocated treatment Aspirin 0.87 (0.67-1.14) 0.30 0.89 (0.68-1.16) 0.37 Aspirin + dipyridamole 0.78 (0.59-1.03) 0.078 0.75 (0.57-0.99 0.041 Vitamin K antagonists Reference Reference Age 56-62 years 1.2 (0.90-1.6) 0.21 1.2 (0.90-1.6) 0.22 > 62 years 1.6 (1.2-2.1) <0.001 1.4 (1.1-1.9) 0.009 up to 55 years Reference Reference Sex Female 1.2 (0.86-1.6) 0.31 - Smoking 1.3 (0.97-1.9) 0.079 - NYHA Class III or IV 1.4 (1.1-1.8) 0.018 1.4 (1.1-1.8) 0.018 Diabetes Mellitus 1.2 (0.82-1.8) 0.35 - Body Mass Index (kg/m2) 0.99 (0.95-1.03) 0.70 - Previous ATE (AMI or stroke) 1.2 (0.93-1.5) 0.187 - Hyperlipidaemia 0.83 (0.66-1.05) 0.121 - Hypertension 0.99 (0.78-1.3) 0.92 - Type of graft usedVein graft only 1.4 (1.1-1.7) 0.008 1.3 (1.1-1.8) 0.018 IMA + vein grafts Reference Reference Vessel disease Triple vessel 1.03 (0.46-2.3) 0.94 - Two vessel 1.1 (0.49-2.5) 0.81 - Single vessel Reference Digitalis 3.1 (1.8-5.6) <0.001 2.3 (1.3-4.3) 0.006 Lipid lowering drugs 0.56 (0.35-0.87) 0.011 0.55 (0.35-0.87) 0.011 Beta blocking agents 0.70 (0.54-0.90) 0.005 - Diuretics 1.3 (0.96-1.7) 0.087 -

0.10.2 0.5 1.0 2 5 10.0

Figure 2 Univariate and multivariable results of Cox proportional hazards regression analysis of MACE.

For myocardial infarction, which had occurred in 177 patients (82 fatal), the cumulative incidence over the 14-year period was 31%. In the patients treated with aspirin, aspirin with dipyridamole and VKA, this was 33%, 29% and 30%, again not significantly different (HR= 1.1 (95%CI, 0.80 to 1.6 P=0.46) and HR=0.96 (95%CI, 0.66 to 1.4; P=0.82) for aspirin and aspirin with dipyridamole, compared to VKA, P=0.60 (aspirin compared to aspirin with dipyridamole HR=1.2, 95%CI, 0.83 to 1.7; P=0.34); see Figure 3).

- 101 - CABADAS 14-year follow-up study

Repeat revascularization was performed in 106 patients (in 76 patients as the only MACE event), resulting in a cumulative incidence at 14 years of 23%. Of these, 72% was a percutaneous coronary intervention and 28% repeat coronary artery bypass surgery (comparable between aspirin, aspirin with dipyridamole and VKA; p=0.30).

Cardiac death Myocardial infarction Repeat revascularization 40% 40% 40% Aspirin Aspirin Aspirin Aspirin + dipyridamole Aspirin + dipyridamole Aspirin + dipyridamole 30% VKA 30% VKA 30% VKA

20% 20% 20%

10% 10% 10% P=0.96 P=0.60 P=0.006 0% 0% 0% 02468101214 02468101214 02468101214 Time (years) Time (years) Time (years) Figure 3 Kaplan Meier of cardiac death, myocardial infarction and repeat revascularization after coronary-artery-bypass-graft surgery, separately.

With 35% at 14 years, the need for repeat revascularization was most pronounced in patients treated with VKA, compared to 17% in patients treated with aspirin and 18% in patients treated with aspirin with dipyridamole (HR= 2.2 (95%CI, 1.4 to 3.5; P=0.001) and HR=2.2 (95%CI, 1.3 to 3.5; P=0.002), respectively, p=0.006 (aspirin compared to aspirin with dipyridamole HR=0.99, 95%CI, 0.57 to 1.7; P=0.97); see Figure 3). Furthermore, repeat revascularization was most frequently performed in younger patients below 56 years of age at the time of surgery, with a 2.8- fold increased risk of repeat revascularization compared to patients aged above 62 years (95%CI, 1.6 to 4.8, P=0.001). When comparing the patients aged 56-62 years with patients above 62 years, a 1.8-fold increased risk was observed (95%CI, 1.03 to 3.3, P=0.039). Additional risk-indicators were NYHA class III or IV (HR=2.0, 95%CI, 1.2 to 3.2; p=0.007), Use of digitalis (HR=5.3, 95%CI, 1.6 to 17.8;p=0.006), both prior

to surgery and Number of grafts used during surgery (HR 1 or 2 grafts vs. 3 or more = 1.8, 95%CI, 1.1 to 2.8; p=0.018).

- 102 - Chapter 6

Continuation of allocated treatment After one year, when the CABADAS study had ended, continuation of the allocated treatment was left at the discretion of the treating cardiologist. As presented in Table 2, at the first extended follow-up evaluation, after on average 2.5 years (SD, 0.8), the majority of the patients originally randomized to aspirin continued with aspirin (78%, 13% in combination with dipyridamole). Dipyridamole alone was used by 1%, 4% had switched to VKA and 17 % had stopped without switching. In the patients originally randomized to aspirin with dipyridamole, aspirin was still used by 77% of the patients (31% still in combination with dipyridamole), dipyridamole alone by 2%, 4% had switched to VKA and 17% had stopped without switching. Almost half of the patients originally randomized to VKA were still using VKA (49%), 30% had switched to aspirin (3% combined with dipyridamole) and 21% of patients had stopped without switching to another drug.

Table 2 Continuation of allocated treatment.

Aspirin Aspirin + VKA† Total P-value Dipyridamole N=248 N=234 N=244 N=726

Medication at 2.5 years Anticoagulant therapy, % Aspirin 65 46 27 46 <0.01 Aspirin with dipyridamole 13 31 3 16 Dipyridamole 1 2 0 1 Vitamin K antagonists 4 4 49 19 None 17 17 21 18

Concomitant drugs, % beta blockers 31 29 36 32 0.25 calcium antagonists 10 11 15 12 0.23 nitrates 9 7 10 9 0.60 diuretics 14 12 15 14 0.58 lipid lowering drugs 25 30 28 27 0.46 ACE inhibitor 12 14 15 14 0.60 digitalis 4 4 3 4 0.97

- 103 - CABADAS 14-year follow-up study

Regarding concomitant cardiovascular medication, in all three treatment groups, particularly the use of lipid lowering drugs and ACE inhibitors had increased from 5% to 27% and 3% to 14% respectively. Furthermore, the use of nitrates had increased from 1% to 9%, consistent with the presence of anginal complaints at the first extended follow-up evaluation, i.e. 89% NYHA I, 10% NYHA II and 1% NYHA III. When comparing the use of concomitant medication in the patients originally treated with aspirin, aspirin with dipyridamole and VKA, no apparent differences were observed.

Sensitivity analysis In addition to the primary “Intention-To-Treat” analysis, a sensitivity analysis was performed using an “As Treated” approach. In this analysis, the continuation of allocated treatment was taken into account by using a varying exposure model in which treatment was allowed to change according to actual treatment after the time of the first extended follow-up. Overall, for MACE, differences between treatments were non-significant (P=0.20). Furthermore, the hazard ratio’s from this analysis were comparable to our “Intention-To-Treat” findings. Again, VKA was associated with a higher risk for MACE than aspirin with dipyridamole and aspirin (HR=1.4, 95%CI, 1.04 to 2.0; P=0.030 and HR=1.2, 95%CI, 0.94 to 1.6; P=0.136, respectively. When comparing VKA with no treatment, VKA patients were at 1.2 higher risk for MACE (95%CI, 0.76 to 1.7; P=0.51). Aspirin with dipyridamole and aspirin did not differ from those that stopped. In reference to repeat revascularization, also in the varying exposure approach, the risk was higher in VKA, compared to aspirin with dipyridamole and aspirin, as well as to stopping treatment (HR=3.1, 95%CI, 1.7 to 5.7; P=0.002 and HR=2.2, 95%CI, 1.4 to 3.4; P=0.001 and HR=2.4, 95%CI, 1.3 to 4.7; P=0.007, respectively). In this, aspirin with dipyridamole, aspirin and no treatment were not significantly different.

COMMENT

We evaluated the 14-year clinical outcome of patients who underwent coronary artery bypass grafting and subsequent treatment with aspirin, aspirin with dipyridamol or VKA. Studies with such a long-term follow-up are limited. In our study, cumulative rates of major adverse cardiac events reached 52%, with 17% at 5 year, 34% at 10 year

- 104 - Chapter 6

and 41% at 12 year. When taking differences in follow-up into account, these rates are comparable with those from previous reports [9-12]. Overall, with 49%, 50% and 59% at 14-year, occurrence of MACE did not differ between the three treatment groups. When evaluating the three components of MACE separately, the risk of both cardiac mortality and myocardial infarction was comparable in patients treated with aspirin, aspirin with dipyridamol and VKA. For repeat revascularization, however, a significant difference in favor of aspirin and aspirin plus dypiridamole was observed. Patients treated with VKA had a 2-fold increased risk for repeat revascularization. In this, patients treated with VKA already started to differ from aspirin and aspirin with dipyridamol early after completion of the original CABADAS one-year follow-up.

40% Optimal VKA Non-optimal VKA 30%

20%

10% Repeat revascularization P = 0.63 0% 02468101214 Time (years) Figure 4 Kaplan Meier of repeat revascularization after coronary-artery-bypass-graft surgery, in patients treated with VKA.

Although differences in the separate components of MACE are well recognized [11,13-15], in our study this finding is intriguing. Firstly, one-year angiographic results from the CABADAS study showed comparable graft occlusion in the three treatment groups, which is not consistent with the observed difference already present within the second year of follow-up [8,16]. Secondly, we confirmed this finding in a secondary analysis using a varying exposure approach, taking into account the changes made to the randomized study treatment. Thirdly, when evaluating the quality of VKA therapy (i.e. percentage of time spend below the target range and the stability of INR)

- 105 - CABADAS 14-year follow-up study

during the first year, no difference in repeat revascularization was observed in patients with optimal VKA and patients with a non-optimal VKA (additional evaluation, see Figure 4) These three observations induced us to speculate that the high rate of repeat revascularization in patients treated with VKA is in part due to a referral bias. VKA patients were seen every two to three weeks by the specialized thrombosis service for the monitoring of the level of anticoagulation, whereas patients treated with aspirin or aspirin with dipyridamole were seen at a 3-montly interval only during the first year, and thereafter only when clinically indicated. The frequent contacts in VKA patients and continued care provided by the specialized thrombosis service might have increased sensitivity towards referral to cardiac specialists, resulting in a lower threshold towards coronary angiography and an increased probability for repeat revascularization. This could explain the difference between repeat revascularisation and the “hard” clinical outcome events of cardiac death and myocardial infarction. Repeat revascularisation might be less suitable as a study endpoint in the setting of an observational follow-up study particularly when there are inherent differences in the structure of care associated with the treatments that are evaluated. In this setting, the “hard” clinical outcome events cardiac death and myocardial infarction should be used for valid risk estimates associated with the different treatments. In our study, using these endpoints no difference between aspirin, aspirin with dipyridamole and VKA was observed. Although CABADAS was a randomized controlled clinical trial, a major limitation of our study is the retrospective setting of the extended follow-up. As already mentioned, referral bias might have lead to the observed difference in repeat revascularization. A prospective design with visits at regular intervals in all patients would have reduced this bias. Furthermore, treatment and clinical outcome were evaluated at the first extended follow-up at 2.5 years. After this first follow-up, occurrence of MACE was thoroughly assessed but detailed information on continuation of medical treatments after 2.5 years is no longer available, limiting the ability to further assess the differences in repeat revascularization rates. As the risk of both cardiac mortality and myocardial infarction did not differ between the three treatments groups, and irrespective of the extend to which referral

- 106 - Chapter 6

bias might have led to the increased rate of repeat revascularization in patients treated with VKA, our long-term results add further evidence to the recommendations for antithrombotic therapy in CABG patients, as recently re-established in the 8th edition of the American College of Chest Physicians evidence-Based Clinical Practice Guidelines [2]. For all patients with coronary artery disease undergoing CABG, aspirin indefinitely is recommended and as there is no evidence that adding dipyridamole to aspirin improves outcome, this is not recommended. VKA should only be considered if indicated because of concomitant condition, e.g. heart valve replacement. In those instances, VKA combined with aspirin is suggested [2]. In conclusion, this study with 14-year clinical outcome provide further evidence for the use of aspirin as secondary prophylaxis after coronary-artery-bypass-graft surgery.

- 107 - CABADAS 14-year follow-up study

References

1. Antiplatelet Trialists' Collaboration. Collaborative overview of randomised trials of antiplatelet therapy--II: Maintenance of vascular graft or arterial patency by antiplatelet therapy. BMJ 1994; 308: 159-168. 2. Becker RC, Meade TW, Berger PB, Ezekowitz M, O'Connor CM, Vorchheimer DA, Guyatt GH, Mark DB, Harrington RA. The primary and secondary prevention of coronary artery disease: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest 2008; 133: 776S-814S. 3. Eagle KA, Guyton RA, Davidoff R, Edwards FH, Ewy GA, Gardner TJ, Hart JC, Herrmann HC, Hillis LD, Hutter AM, Jr., Lytle BW, Marlow RA, Nugent WC, Orszulak TA, Antman EM, Smith SC, Jr., Alpert JS, Anderson JL, Faxon DP, Fuster V, Gibbons RJ, Gregoratos G, Halperin JL, Hiratzka LF, Hunt SA, Jacobs AK, Ornato JP. 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). J Am Coll Cardiol 2004; 44: e213-e310. 4. Chesebro JH, Fuster V, Elveback LR, Clements IP, Smith HC, Holmes DR, Jr., Bardsley WT, Pluth JR, Wallace RB, Puga FJ, et al. Effect of dipyridamole and aspirin on late vein-graft patency after coronary bypass operations. N Engl J Med 1984; 310: 209-214. 5. Lorenz RL, Schacky CV, Weber M, Meister W, Kotzur J, Reichardt B, Theisen K, Weber PC. Improved aortocoronary bypass patency by low-dose aspirin (100 mg daily). Effects on platelet aggregation and thromboxane formation. Lancet 1984; 1: 1261-1264. 6. Mangano DT. Aspirin and mortality from coronary bypass surgery. N Engl J Med 2002; 347: 1309- 1317. 7. Sharma GV, Khuri SF, Josa M, Folland ED, Parisi AF. The effect of antiplatelet therapy on saphenous vein coronary artery bypass graft patency. Circulation 1983; 68: II218-II221. 8. van der Meer J, Hillege HL, Kootstra GJ, Ascoop CA, Mulder BJ, Pfisterer M, van Gilst WH, Lie KI. Prevention of one-year vein-graft occlusion after aortocoronary-bypass surgery: a comparison of low-dose aspirin, low-dose aspirin plus dipyridamole, and oral anticoagulants. The CABADAS Research Group of the Interuniversity Cardiology Institute of The Netherlands. Lancet 1993; 342: 257-264. 9. Cameron AA, Green GE, Brogno DA, Thornton J. Internal thoracic artery grafts: 20-year clinical follow-up. J Am Coll Cardiol 1995; 25: 188-192. 10. Daemen J, Boersma E, Flather M, Booth J, Stables R, Rodriguez A, Rodriguez-Granillo G, Hueb WA, Lemos PA, Serruys PW. Long-term safety and efficacy of percutaneous coronary intervention with stenting and coronary artery bypass surgery for multivessel coronary artery disease: a meta-analysis with 5-year patient-level data from the ARTS, ERACI-II, MASS-II, and SoS trials. Circulation 2008; 118: 1146-1154. 11. The BARI Investigators. The final 10-year follow-up results from the BARI randomized trial. J Am Coll Cardiol 2007; 49: 1600-1606. 12. van Brussel BL, Voors AA, Ernst JM, Knaepen PJ, Plokker HW. Venous coronary artery bypass surgery: a more than 20-year follow-up study. Eur Heart J 2003; 24: 927-936.

- 108 - Chapter 6

13. The Bypass Angioplasty Revascularization Investigation (BARI) Investigators. Comparison of coronary bypass surgery with angioplasty in patients with multivessel disease. N Engl J Med 1996; 335: 217-225. 14. 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 coronary-artery bypass surgery and stenting for the treatment of multivessel disease. N Engl J Med 2001; 344: 1117-1124. 15. Serruys PW, Ong AT, van Herwerden LA, Sousa JE, Jatene A, Bonnier JJ, Schonberger JP, Buller N, Bonser R, Disco C, Backx B, Hugenholtz PG, Firth BG, Unger F. Five-year outcomes after coronary stenting versus bypass surgery for the treatment of multivessel disease: the final analysis of the Arterial Revascularization Therapies Study (ARTS) randomized trial. J Am Coll Cardiol 2005; 46: 575-581. 16. van der Meer J, Brutel de la Riviere A, van Gilst WH, Hillege HL, Pfisterer M, Kootstra GJ, Dunselman PH, Mulder BJ, Lie KI. Effects of low dose aspirin (50 mg/day), low dose aspirin plus dipyridamole, and oral anticoagulant agents after internal mammary artery bypass grafting: patency and clinical outcome at 1 year. CABADAS Research Group of the Interuniversity Cardiology Institute of The Netherlands. Prevention of Coronary Artery Bypass Graft Occlusion by Aspirin, Dipyridamole and Acenocoumarol/Phenprocoumon Study. J Am Coll Cardiol 1994; 24: 1181-1188.

- 109 -