By Carrier, M Emery, R; Kandzari, D E; Harrington, R; Guertin, M-C; Tardif, J-C
Aim. Aortic cross-clamp time remains a significant marker of mortality and morbidity after coronary artery bypass graft (CABG) surgery. Pyridoxal-5-phosphate (MC-1), blocking purinergic receptors and intracellular influx of calcium, was shown to decrease the incidence of perioperative myocardial infarction in the prospective, randomized, double-blinded MC-1 to Eliminate Necrosis and Damage in CABG (MEND-CABG) clinical trial. Methods. We studied the relationship between treatment with MC-1 and aortic cross-clamping relative to the incidence of cardiovascular (CV) death and myocardial infarction (MI) in the trial that enrolled 901 high-risk patients undergoing CABG with cardiopulmonary bypass. Patients were randomized to receive either placebo, MC-1 250 mg/day or MC-1 750 mg/ day starting 3-10 h before CABG and continued for 30 days after surgery. Serial creatine kinase-myocardial band (CK-MB) determinations, ECGs and clinical evaluations were performed.
Results. Cross-clamping time increased the event rate of death and MI with an odds ratio (95% confidence interval) of 1.67 (1.17- 2.37, P=0.0044). Treatment with MC-1 decreased the rate of events (P=0.0073) with odds ratios of 0.52 (0.31-0.88 for MC-1 250 mg/day versus placebo) and 0.48 (0.29-0.82 for MC-1 750 mg/day versus placebo). There was no interaction between cross-clamp time and treatment (P=0.61) on the occurrence of the combined endpoint.
Conclusion. MC-1 decreased the incidence of CV death and MI (CK- MB >/=100 ng/mL) during the first 90 days after CABG in the MEND- CABG trial. Although longer aortic clamping time increased the risk of cardiovascular events, the protective effect of MC-1 was independent of ischemic time during CABG.
KEY WORDS: Ischemia – Coronary artery disease – Coronary artery bypass.
Myocardial ischemia occurring during coronary artery bypass grafting surgery (CABG) remains related with aortic cross clamping and cardioplegic arrest of the heart. Clinical studies have shown that the duration of myocardial ischemia represented by the aortic cross-clamping time was strongly related with early mortality and morbidity after CABG.1,2
Several methods of myocardial protection have been well studied, described in great detail and widely applied in the current clinical practice of cardiac surgery3-5 Pyridoxal-5-phosphate monohydrate, referred to as MC-1, is a natural metabolite of pyridoxine that inhibits purinergic (P2) receptors and the intracellular influx of calcium thereby reducing cell damage in laboratory experiments. In a phase 2 clinical trial, 10 mg/kg/day of MC-1 given for up to 14 days in patients undergoing high-risk percutaneous coronary intervention resulted in a decrease in infarct size over the first 24 h following angioplasty suggesting the cardioprotective properties of MC-1 in the setting of myocardial ischemia and reperfusion.6 The MC-1 to Eliminate Necrosis and Damage in CABG (MENDCABG) trial enrolled 901 high-risk patients who underwent CABG with cardiopulmonary bypass. Patients were randomized to receive either placebo, MC-1 250 mg/day or MC-1 750 mg/day starting 3-10 h before CABG and continued for 30 days after surgery. The study showed that MC-1 may be cardioprotective, especially in preventing larger more significant myocardial infarctions (MI) defined as a postoperative peak CK-MB greater than 100 ng/mL.7
The objective of the present analysis was to retrospectively evaluate the effects of treatment with MC1 and ischemia associated with aortic cross-clamping on the incidence of cardiovascular (CV) death and MI in the MEND-CABG trial.
Materials and methods
Study protocol and patient population
The patient population, study design and main findings of the MEND-CABG trial have been published.7 In summary, eligible patients included those scheduled for isolated CABG under cardiopulmonary bypass with at least two of the following baseline risks factors: age >/=65 years; current smoker; history of diabetes mellitus requiring treatment other than diet; history of non-disabling stroke, transient ischemic attack or carotid endarterectomy; prior peripheral artery surgery or angioplasty; recent MI between 48 h and 6 weeks before CABG; evidence of left ventricular dysfunction or congestive heart failure; need for urgent CABG; renal dysfunction with creatinine between 133 and less than 250 [degrees]=mol/L (1.5- 2.8 mg/dL); or presence of stenosis (>50%) in at least one carotid artery. The trial was conducted in 40 clinical centers in Canada and the United States, and the institutional review boards or equivalent approved the protocol at each site. All patients provided written informed consent.
Patients were randomized in a double-blind fashion to receive MC- 1 250 mg, MC-1 750 mg, or matching placebo tablets administered orally once daily. The first dose of study medication was administered 3-10 h prior to CABG surgery and treatment continued for 30 days after surgery. Patients were followed for in-hospital adverse events and clinical endpoints. In addition, patients underwent clinical evaluation at 30 and 90 days after CABG surgery.
Endpoints
For the present study focusing on myocardial ischemia and aortic cross clamping time during CABG surgery, CV death and MI were analyzed 90 days after surgery. Death was considered of cardiovascular etiology unless there was a clear non-cardiovascular cause. Myocardial infarction was defined as follows in our post-hoc analysis, based on central laboratory or core electrocardiogram (ECG) laboratory data: a peak creatinine kinase-myocardial band (CK- MB) of 100 ng/mL or above on days up to and including postoperative (POD) 4; or a new Q-wave evidence of MI along with CK-MB of 35 ng/ mL or above on days up to and including POD 4; or a peak CK-MB of 25 ng/mL or above occurring after POD 4; or a new Q-wave evidence of MI that was not present at POD 4; or a Q-wave or non-Q-wave MI as identified by the investigator.
Patient blood samples were collected before CABG surgery (baseline) and 4, 8, 12, 16, 24, 36, 48, 72 and 96 h and 30 and 90 days after CABG surgery. All samples were submitted to a central laboratory (Cirion BioPharm Research, Inc., Laval, QC, Canada) for CKMB measurements. A 12-lead ECG was performed at baseline and on POD 2, 4, 30 and 90 and submitted to the core laboratory (Dynacare Laboratories, Edmonton, AB, Canada) for centralized reading. Electrocardiographic tracings were analyzed using the Minnesota code.
Statistical analysis
The present study focused solely on the 861 patients in whom the aortic cross clamping time was available (861/901, 95%). Multivariate logistic regressions were used to study the effects of MC-I treatment and aortic cross-clamping time on the combined endpoint of CV death and MI as previously defined. The interaction between aortic cross-clamping time and treatment was investigated. Logistic regression models were also used to adjust for potential covariates such as type of cardioplegic solutions used (blood or crystalloid) and the baseline risk factors identified above. Baseline risk factors retained to be part of the final model are those that showed an association (P
TABLE I.-Patient characteristics.
All statistical testing was performed at a two-sided 0.05 significance level. Analyses were done with SAS 8.2.
Results
Patient characteristics, operative data and clinical outcomes
The mean patient age was 65+-10 years (range, 34 to 95 years), and 79% of patients were men (Table I). An average of 3.5+-1.0 coronary bypass grafts was performed per patient (Table II). Most patients underwent cardioplegic arrest of the heart with a blood based solution administered at cold or warm temperature (784/842, 93%), compared with a crystalloid solution (58/842, 7%). The type of solution used was not known in 19 patients. Among the 861 patients who were available for the analysis, 97 had an event (CV death or MI up to day 90 after CABG, 97/861, 11%) (Table III).
Effects of aortic cross-clamping time, MC-1 and other factors on clinical outcomes after CABG
Aortic cross-clamping time averaged 66+-31 min in the 861 patients included in the present analysis. Crossclamping time increased the event rate of CV death and MI with an odds ratio (OR) (95% confidence interval) of 1.67 (1.17-2.37, P=0.0044). Treatment with MC-1 decreased the rate of events (P=0.0073) with ORs of 0.52 (0.31-0.88 for MC-1 250 mg/day versus placebo) and 0.48 (0.29-0.82 for MC-1 750 mg/day versus placebo). There was no interaction between cross-clamp time and treatment (P=0.61) on the occurrence of the combined endpoint.
TABLE II.-Operative data.
TABLE III.-Clinical outcomes at 90 dayspost-CABG.
After adjusting for the type of cardioplegic solution and the risk factors found to be associated with the combined endpoint in the univariate analyses, namely renal dysfunction and carotid stenosis, aortic cross clamping time was still associated with more MIs and CV deaths after CABG (OR: 1.65, 1.13-2.41, P=0.0098) and MC- 1 treatment significantly (P=0.0088) decreased the event rates (OR: 0.53, 0.31-0.89 and OR: 0.48, 0.28-0.82 respectively for MC-1 250 mg and 750 mg). Again, no interaction was found between cross-clamp time and treatment (P=0.77).Effect of prolonged aortic cross clamping time on clinical outcomes Among the 448 patients with aortic cross-clamping times longer than 60 min, 58 patients (58/ 448, 12.9%) had a MI or CV death after CABG, compared with 39 who had events of the 413 patients (39/413, 9.4%) who underwent shorter cross-clamping. Aortic cross-clamping time longer than 60 min increased event rates with an OR of 1.43 (0.93-2.20, P=0.1059). Treatment with MC-1 250 mg/day and 750 mg/day decreased event rates (P=0.0085) with ORs of 0.52 (0.31-0.87) and 0.50 (0.30-0.84) respectively. There was no interaction between treatment with MC-1 and aortic cross clamping time of 60 min (P=0.865) (Table IV).
TABLE IV.-Effect of aortic cross-clamping time on composite endpoint of death and MI.
Discussion
The MEND-CABG trial demonstrated that MC-1, a purinergic P2 receptor inhibitor, was effective in reducing the incidence of perioperative MI defined by peak CK-MB values greater that 100 ng/ mL after CABG surgery. Because several studies have clearly demonstrated that elevated CK-MB release after CABG was strongly associated with worse 6-month and one-year survival rates, MC-1 could become an interesting adjunct treatment for CABG patients.7,8
Consistent with previous studies, this trial demonstrated a direct relationship between cross-clamp time and the extent of myocardial injury. The risk of myocardial infarction after CABG was higher in patients with cross-clamping times of 60 min or more, and was almost doubled in patients with an aortic cross-clamping time equal to or longer than 90 min. In addition, MC-I was effective in decreasing by almost half the risk of myocardial infarction (CK-MB >100 ng/mL) after CABG, and there was no statistical interaction between MC-1 treatment and aortic cross-clamping time, suggesting that MC-1 is as effective in patients with long ischemic times as it is for those with shorter cross-clamping.
The duration of aortic cross-clamping has previously been shown to be a predictor of clinical outcomes after CABG. Our study confirms this relationship in a population at high risk for events, particularly if we consider the incidence of MI in the placebo group. This relationship persisted when the results were adjusted for different factors including the type of cardioplegic solution. The present trial was not designed to study the role of different solutions used for cardioplegic arrest. Although Mallidi et all suggested that warm or tepid blood cardioplegic arrest may be associated with better early and late event-free survival than is cold cardioplegia, the present analysis showed that myocardial ischemic injury remains of serious concern despite the use of blood cardioplegic arrest during CABG surgery. The adjunct treatment with MC-1 could increase myocardial resistance to injury during cardioplegic arrest with ischemia and at the time of myocardial reperfusion, which is an unmet clinical need. MC-1 is a purinergic P2-receptor antagonist that inhibits the binding of ATP and thereby decreases calcium flux into cardiomyocytes.9, 10 Extracellular ATP is released during various stress conditions, such as ischemia or reperfusion injury, and acts on purinergic P2-receptors to initiate an increase in intracellular calcium. Because calcium overload leads to cell damage and is responsible for cell death during ischemia followed or not by reperfusion, reduction in calcium influx may explain the beneficial effects observed with MC-1 in animal models and in clinical studies.
Several large clinical trials have examined therapies intended to decrease myocardial ischemic damage in the peri-operative setting. The GUARDIAN trial tested the effect of cariporide, a sodium/ hydrogen exchanger inhibitor, and yielded overall negative results.11, 12 The PRIMO-CABG I trial in testing the effect of pexelizumab, a C5 complement inhibitor, resulted in a reduction in death and MI events after CABG that did not reach statistical significance.13 Similarly, the PRIMOCABG II trial enrolled a larger group of patients and was also unable to show a statistically significant reduction in the primary end-points combining post- operative death and MI. An effective approach to reduce perioperative events in high-risk patients undergoing CABG is therefore required.
The reduction in CV death and MI with MC-1 in the MEND-CABG trial was observed when a cut-off value of CK-MB >/=100 ng/mL was used in a post-hoc analysis. The reduction in the prespecified composite endpoint did not reach statistical significance in the main study when the cut-off value of 50 ng/mL was used.7 Although the post-hoc nature of this finding represents a limitation, the cut-off value of CK-MB of 100 ng/mL was explored because of previous studies linking it with future cardiovascular events.14 The PRIMO-CABG trial using a similar cut-off value of CK-MB of at least 100 ng/mL after surgery showed that patients with adjudicated MI through day 4 experienced a higher rate of mortality up to 180 days after surgery.13 The GUARDIAN trial also showed that patients with CK-MB values greater than 10 times the upper limit of normal after CABG had a higher rate of mortality 6 months after surgery.14 The two latter clinical trials suggest that large amounts of CK-MB enzyme release after CABG are associated with significant elevations of medium-term mortality after surgery.
Conclusions
MC-1 decreased the incidence of myocardial infarction (>100 ng/ mL) and death during the first 90 days after CABG in the MEND-CABG trial. Although longer aortic cross-clamping time increased the risk of cardiovascular events, the protection with MC-I was independent of ischemic time during CABG. The adjunct treatment with MC-I would therefore be expected to benefit high-risk patients undergoing CABG, irrespective of the aortic cross-clamping time and even after adjustment for type of administration of cardioplegic solutions.
Fundings.-The MEND-CABG trial was funded by Medicare, Winnipeg, Canada.
Acknowledgements.-Statistical analysis was done by the Montreal Heart Institute Coordinating Center.
Received on January 14, 2008.
Accepted for publication on January 17, 2008.
References
1. Raman JS, Bellomo R, Hayhoe M, Tsamitros M, Buxton BF. Metabolic changes and myocardial injury during cardioplegia: a pilot study. Ann Thorac Surg 2001;72:1566-71.
2. Onorati F, De Feo M, Mastroroberto P, Cristodoro L, Pezzo F, Renzulli A et al. Determinants and prognosis of myocardial damage after coronary artery bypass grafting. Ann Thorac Surg 2005;79:837- 45.
3. Mallidi HR, Sever J, Tamariz M, Singh S, Hanayama N, Christakis GT et al. The short-term and long-term effects of warm or tepid cardioplegia. J Thorac Cardiovasc Surg 2003;125:711-20.
4. Rosenkranz ER, Buckberg GD, Laks H, Mulder DG. Warm induction of cardioplegia with gluthamate-enriched blood in coronary patients with cardiogenic shock who are dependent on inotropic drugs and intra-aortic balloon support. J Thorac Cardiovasc Surg 1983;86:507- 18.
5. Buckberg GD. Myocardial protection: entering the new millennium. J Card Surg 2002;17:447-50.
6. Kandzari DE, Labinaz M, Cantor WJ, Madan M, Gallup DS, Hasselblad V et al. Reduction of myocardial ischemic injury following coronary intervention (the MC-I to Eliminate Necrosis and Damage trial). Am J Cardiol 2003;92:660-4.
7. Tardif JC, Carrier M, Kandzari DE, Emery R, Cote R, Heinonen T et al. for MEND-CABG Investigators. Effect of pyridoxal-5-phosphate in patients undergoing high-risk coronary artery bypass surgery: results of the MEND-CABG study. J Thorac Cardiovasc Surg 2007; 133:1604-11.
8. Klatte K, Chaitman BR, Theroux P, Gavard JA, Stocke K, Boyce S et al. for the GUARDIAN Investigators. Increased mortality after coronary artery bypass graft surgery is associated with increased levels of postoperative creatine kinase-myocardial band isoenzyme release. J Am Coll Cardiol 2001;38:1070-7.
9. Wang X, Dakshinamurti K, Musat S, Dhalla NS. Pyridoxal 5- phosphate is an ATP-receptor antagonist in freshly isolated rat cardiomyocytes. J MoI Cell Cardiol 1999;31:1063-72.
10. Trezise DJ, Bell NJ, Khakh BS, Michel AD, Humphrey PA. P2 purinoceptor antagonist properties of pyridoxal-5-phosphate. Eur J Pharmacol 1994;259:295-300.
11. Theroux P, Chaitman BR, Danchin N, Erhardt L, Meinertz T, Schroeder JS et at. Inhibition of the sodium-hydrogen exchanger with cariporide to prevent myocardial infarction in high-risk ischemie situations. Main results of the GUARDIAN trial. Guard during ischemia against necrosis (GUARDIAN) investigators. Circulation 2000; 102:3032-8.
12. Chaitman BR. A review of the GUARDIAN trial results: clinical implications and the significance of elevated perioperative CKMB on 6 months survival. J Card Surg 2003;1:13-20.
13. Verrier ED, Shernan SK, Taylor KM, Van de Werf F, Newman MF, Chen JC, PRIMO-CABG Investigators. Terminal complement blockade with pexelizumab during coronary artery bypass graft surgery requiring cardiopulmonary bypass: a randomized trial. JAMA 2004;291:2319-27.
14. Gavard JA, Chaitman BR, Sakai S, Stocke K, Danchin N, Erhardt L etal. for the GUARDIAN Investigators. Prognostic significance of elevated creatine kinase MB after coronary bypass surgery and after an acute coronary syndrome. J Thorac Cardiovasc Surg 2003; 126: 807- 13.
M. CARRIER 1, R. EMERY 2, D. E. KANDZARI3, R. HARRINGTON 3, M.- C. GUERTIN 4, J.-C. TARDIF 5
1 Department of Surgery
Montreal Heart Institute and Universite de Montreal
Montreal, QC, Canada
2 Department of Cardiovascular Surgery
St. Joseph’s Hospital Health East Care System
St. Paul, MN, USA
3 Department of Interventional Cardiology Research
Duke Clinical Research Institute
Durham, NC, USA
4 Montreal Heart Institute Coordinating Center
Montreal, QC, Canada
5 Department of Medicine
Montreal Heart Institute and Universite de Montreal
Montreal, QC, Canada
Address reprint requests to: M. Carrier, Department of Surgery, Montreal Heart Institute, 5000 Belanger Street, Montreal, QC, H1T 1C8, Canada. E-mail: [email protected]
Copyright Edizioni Minerva Medica Apr 2008
(c) 2008 Journal of Cardiovascular Surgery. Provided by ProQuest Information and Learning. All rights Reserved.
Comments