| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL ARTICLE |
Division of Cardiothoracic, Surgery Siriraj Hospital, Faculty of Medicine, Bangkok, Thailand
Sanya Thiengburanatham, MD Tel: +66 815322147 Fax: +66 22443282 Email: sanya_theing{at}yahoo.com, Division of Cardiothoracic Surgery, Department of Surgery, Bangkok Metropolitan Administration Medical College and Vajira Hospital, Bangkok, Thailand.
ABSTRACT
Hyperhomocysteinemia and other major cardiovascular risk factors are associated with increased vascular oxidative stress. To access the effects preoperative plasma homocysteine levels and other atherosclerotic risk factors on myocardial ischemia-reperfusion injury after conventional coronary artery bypass, 213 patients with normal renal function were enrolled prospectively. Cardiac troponin T was measured postoperatively to determine myocardial injury. There was a significant relationship between hyperhomocysteinemia and postoperative peak troponin T. This was more marked in patients without major atherosclerotic risk factors than in those who had at least one risk factor. Moreover, among current cigarette smokers, those with the highest preoperative plasma homocysteine levels had the lowest postoperative troponin T levels. From multivariate linear regression analysis, the predictors of high postoperative troponin T were hyperhomocysteinemia, hypertension, and aortic crossclamp time, but the presence of major atherosclerotic risk factors paradoxically modified the effects of hyperhomocysteinemia on postoperative myocardial ischemia-reperfusion injury.
Key Words: Cardiopulmonary Bypass • Coronary Artery Bypass • Homocysteine • Hyperhomocysteinemia • Myocardial Reperfusion Injury
INTRODUCTION
Conventional coronary artery bypass grafting (CABG), utilizing cardiopulmonary bypass (CPB) with cardioplegic arrest, inevitably elicits myocardial ischemia-reperfusion injury (IRI).1,2 Many of these patients manifest postoperative adverse cardiac events such as myocardial infarction, left ventricular dysfunction, or myocardial failure.3 Two of the known principal mediators of myocardial IRI are intracellular Ca2+ overload and increased oxidative stress due to elaboration of reactive oxygen species.3 Hyperhomocysteinemia (HHcy) has been proposed as an independent risk factor for the development of atherosclerotic disease with subsequent major adverse cardiac and cerebrovascular events and cardiovascular mortality.4–6 Previous studies found evidence of an association between HHcy and myocardial injury after percutaneous coronary interventions and acute coronary syndrome.7,8 One of the proposed mechanisms of HHcy is an increase in vascular oxidative stress leading to endothelial dysfunction.9 In conventional CABG, it is unclear whether HHcy has any association with the extent of myocardial IRI. Moreover, many major atherosclerotic risk factors can also exacerbate reactive oxygen species production. It is well known that diabetes mellitus, hypertension, dyslipidemia, and cigarette smoking are associated with increased vascular oxidant stress and reduced NO bioavailability.10–13 How each of these risk factors interacts with homocysteine (Hcy) in precipitating myocardial IRI is unknown. In this study, we assessed the hypothesis that myocardial IRI after conventional CABG is related to the plasma Hcy level on admission.
PATIENTS AND METHODS
We prospectively studied 213 patients with coronary artery disease undergoing routine CABG at the Siriraj Hospital, Bangkok, Thailand. The study was approved by the hospital ethics committee. Every patient gave written inform consent. Indications for CABG were in accordance with American Heart Association guidelines, 2004. All patients underwent preoperative coronary angiography. Exclusion criteria were emergency operations, additional cardiac procedures, reoperations, recent myocardial infarction (<30 days), and serum creatinine >1.8 mg ·dL–1. Patients with underlying diseases that may affect Hcy levels, such as chronic liver disease, chronic renal failure, thyroid diseases, inflammatory bowel disease, connective tissue disorders (rheumatoid arthritis, systemic lupus erythematosus), and advanced cancer were also excluded. Venous blood samples were obtained from eligible patients on admission. All patients had normal cardiac troponin T (cTnT) and no acute coronary event within 30 days prior to the operation. We defined patients as nonsmokers if they had never smoked tobacco. Ex-smokers were those who had completely ceased smoking for at least 1 year, with history of at least 10 pack-years of cigarette smoking.
Myocardial protection was achieved using antegrade and/or retrograde modified cold blood St. Thomas’ Hospital cardioplegic solution every 20 min after complete diastolic cardioplegic arrest. Topical myocardial hypothermia was accomplished with cold saline. The lowest temperature on CPB was 32°C. All bypass procedures were performed by one of 7 experienced cardiac surgeons. Postoperative cTnT levels immediately on arrival in the intensive care unit (cTnT1) and at 12 h later (cTnT2) were used for assessment of myocardial IRI.14 The cTnT levels were measured by the Elecsys troponin T assay (Elecsys 2010 and Modular Analytics E170; Roche Diagnostics, Mannheim, Germany) using highly specific monoclonal antibodies directed against 2 different epitopes of the cTnT molecule. The coefficients of variation within and between days of the assays were <5%. Blood samples for Hcy levels on admission were taken after at least 8 h of fasting, collected in an EDTA tube, and kept in ice, with immediate transfer to the laboratory. The samples were centrifuged at 3,000 g and the plasma was stored at –80°C pending analysis within 2 weeks. The levels of Hcy were analyzed by fluorescence polarization immunoassay. The coefficients of variation within and between days of the assays were <5%.
Statistical analysis was performed using SPSS version 11.5 software (SPSS, Inc., Chicago, IL, USA). The levels of cTnT and Hcy closely followed a log-normal distribution (correlation 0.99 with expected log-normal values) and were log-transformed before being used as continuous variables. Calculations of odd ratios (OR) with confidence intervals were used to determine the strength of association between cTnT and Hcy quintiles, with cTnT 
0.5 ng ·mL–1 as a cut-off value. Univariate analysis was carried out by determining relationships between individual continuous or dichotomous variables using Spearman’s rho correlation coefficient. The difference in proportion between 2 variables was compared by t test. Any variables with a p value <0.1 were entered into multivariate analysis. The Hcy levels were categorized into quintiles (inter-quintile cut-off points: 9.000, 10.800, 12.000, and 14.51 µmol ·L–1). Group comparisons of the log-transformed values were made by the independent sample t test for 2 groups, and analysis of variance for more than 2 groups, with post-hoc Dunnett corrections. Multivariate linear regression analysis of continuous or dichotomous variables was made using a stepwise linear regression model. All reported probability values were 2-sided, with a p value 
0.05 as the threshold for statistical significance.
RESULTS
Of 213 patients undergoing conventional CABG, 3 had missing values of cTnT2; therefore, analysis of cTnT2 and other variables was performed in 210 cases. Baseline characteristics of the whole study group are shown in Table 1
. There was no difference between the 1st and last Hcy quintiles. There was a significantly higher plasma creatinine level for the 5th Hcy quintile compared to the 1st quintile (p <0.00001), although all creatinine values were within the normal range. Two patients died from postoperative septicemia (all with Hcy in the 1st quintile). The other patients convalesced well without any deterioration of hemodynamic status, and were discharged from the intensive care unit within 3 days after the operation. Figure 1A shows the relationship between cTnT1 and plasma Hcy quintiles. There was no difference in cTnT between the highest and lowest Hcy quintiles (p =0.08, p = 0.494 for 1st vs. 5th Hcy quintiles). There was a significant relationship between cTnT2 and the lowest and 2 uppermost Hcy quintiles, as shown in Figure 1B: p = 0.004, p =0.017 for 1st vs. 4th Hcy quintiles, OR 3.519 (1.35–9.19); p = 0.007 for 1st vs. 5th Hcy quintiles, OR 3.89 (1.50–10.05). Figure 2 shows the relationship between Hcy quintiles among nonsmokers. In Figure 2A, there were no differences in cTnT1 between the 1st and 4th or 5th Hcy quintiles (p =0.061, p = 0.329 for 1st vs. 4th Hcy quintiles; p = 0.277 for 1st vs. 5th Hcy quintiles). However, there were significance differences in cTnT2 between the 2 uppermost Hcy quintiles and the lowest quintiles, as shown in Figure 2B: p =0.001, p = 0.006 for 1st vs. 4th Hcy quintiles, OR 5.077 (1.35–19.17); p = 0.001 for 1st vs. 5th Hcy quintiles, OR 10.5 (2.67–41.29). For ex-smokers and current smokers, no significant differences in either cTnT1 or cTnT2 were found between the 1st and 5th Hcy quintiles (ex-smokers: p = 0.550, p = 0.975 for 1st vs. 5th Hcy quintiles for cTnT1; p =0.093, p = 0.273 for 1st vs. 5th Hcy quintiles for cTnT2; current smokers: p = 0.894, and p = 0.737 for 1st vs. 5th Hcy quintiles for cTnT1, and analysis of variance p =0.276, and p = 0.704 for 1st vs. 5th Hcy quintiles for cTnT2). Therefore, the HHcy patients with history of cigarette smoking did not have more post-CABG myocardial damage than those with normal Hcy levels. Figure 3 shows box plots of the relationship between Hcy quintiles and cTnT2 for each subgroup of preoperative cigarette smoking status. For the 5th Hcy quintile, there was a significantly lower cTnT2 level for current smokers than nonsmokers (p = 0.032, p = 0.019). This result implies that current cigarette smoking might paradoxically reduce the degree of post-CABG myocardial IRI in patients with HHcy. No significant relationship was found in cTnT2 levels among cigarette smoking subgroups in other Hcy quintiles. Table 2 shows the cTnT2 differences between 1st and 5th Hcy quintiles among subgroups with various atherosclerotic risk factors. For patients with no athero-sclerotic risk, there were significance differences in cTnT2 between the 1st and 5th Hcy quintiles. On the other hand, in those with at least one atherosclerotic risk factor, no significance differences were found. There were no differences in cTnT2 between low and high atherosclerotic risk patients among all Hcy quintiles, except for those who had hypertension or high-density lipoprotein (HDL) cholesterol >35 mg ·dL–1. In univariate analysis by Spearman’s rho correlation coefficient (Table 3), there were significant positive correlations between cTnT2 and Hcy quintiles, hypertension status, CPB time, and aortic crossclamp time. HDL cholesterol correlated negatively with cTnT2. Multivariate analysis (Table 4) showed that the predictors of higher cTnT2 values were HHcy (p =0.002), hypertension (p = 0.002), and aortic crossclamp time (p = 0.013). HDL cholesterol was associated with lower cTnT2 by its negative value of standardized coefficients.
|
|
|
|
|
|
|
One of the main outcomes of this study was the significant association between preoperative Hcy levels and the peak value of post-CABG cTnT, which represents the extent of myocardial IRI. Therefore, preoperative HHcy was associated with a higher degree of myocardial IRI in patients undergoing conventional CABG. The threshold of Hcy level on admission for this association was 12 µmol ·L–1, or the 4th Hcy quintile, which is in accordance with previous studies.5 The production of reactive oxygen species is thought to be one of the mechanisms of IRI, which should be augmented by the presence of many atherosclerotic risk factors due to their increase of oxidative stress; however, the results of this study indicate that this might not be the case. The first results of this study might be explained by the higher vascular oxidative stress in HHcy, which could produce more reactive oxygen species and thus greater myocardial IRI.15 It might be possible that in HHcy, endothelial nitric oxide synthase (eNOS) would be uncoupled and produce superoxide anion instead of nitric oxide.16,17 Nitric oxide from the remaining intact eNOS rapidly reacts with superoxide anion to form peroxynitrite, with subsequent endothelial dysfunction and cellular injury.18 One previous study showed a reduction in myocardial damage in patients on CPB after addition of l-arginine to the cardioplegic solution.19 The administration of l-arginine could re-couple eNOS and switch the generation of superoxide anion activity to the production of nitric oxide.
Other interesting results from this study were the negative interactions of various cardiovascular risk factors with HHcy on the degree of myocardial IRI. Many studies have shown that diabetes mellitus, hypertension, and dyslipidemia increase vascular oxidative stress. Nevertheless, these higher oxidative states did not intensify myocardial IRI in the presence of HHcy. Seemingly, the greater myocardial IRI seen in HHcy only affected patients without these cardiovascular risk factors. For patients with hypertension, hypertrophy of the left ventricle per se could jeopardize the subendocardial myocardium during cardioplegic arrest, and hence cause higher cTnT2 levels compared to normotensive patients. The presence of hypertension however did not augment the extent of myocardial IRI in HHcy patients. Also, for nonsmokers, there were even more significant effects of the 2 uppermost Hcy quintiles on the degree of post-CABG myocardial IRI. On the contrary, no significant myocardial IRI was found among patients with history of smoking or current smokers. It is known that cigarette smoking increases reactive oxygen species production, which could theoretically potentiate myocardial IRI.20 Moreover, current cigarette smokers with HHcy had significantly lower cTnT2 values in the 5th quintile compared to nonsmokers.
One possible explanation for these paradoxical results might be found in the balance of eNOS coupling-uncoupling in endothelial dysfunction. HHcy was noted to depend on the presence of coupled eNOS for inducing higher oxidative stress.9 It is known that coupled eNOS normally produces nitric oxide, and uncoupled eNOS produces superoxide anions.18 In patients with high oxidative stress (diabetes, dyslipidemia, and cigarette smoking), much eNOS is in an uncoupled state, and may result in less peroxynitrite being produced than in HHcy patients. Further work is needed to explore peroxynitrite in relation to HHcy.
We concluded that in patients with no major cardiovascular risk factors, HHcy could lead to higher myocardial IRI after CABG, while this effect might be less significant for patients with major atherosclerotic risk factors.
ACKNOWLEDGMENTS
Special thanks to Dr. Somchai Sriyoschati, MD, former Head of the Division of Cardiothoracic Surgery, Siriraj Hospital, for his support and assistance. Thanks to Suthipol Udompunturak, MSc, Institute Statistician, for great help in statistical review and comments. This study was supported by local funding from the Faculty of Medicine, Siriraj Hospital, Bangkok, Thailand.
REFERENCES
Asian Cardiovasc Thorac Ann 2009;
17:483-489
© 2009 by SAGE Publications
DOI: 10.1177/0218492309348635
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
