| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2005;79:1530-1535
© 2005 The Society of Thoracic Surgeons
a Department of Anesthesiology, University Medical Center Utrecht, Utrecht, The Netherlands
b Department of Hematology, University Medical Center Utrecht, Utrecht, The Netherlands
c Department of Thoracic Anesthesiology and Intensive Care, Isala Clinics, Zwolle, the Netherlands
Accepted for publication October 4, 2004.
* Address reprint requests to Dr Lo, Dept of Anesthesiology, University Medical Center Utrecht, PO Box 85500, 3508 GA Utrecht, the Netherlands (E-mail: b.lo{at}azu.nl).
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
METHODS: C-reactive protein was measured in perioperative blood samples of patients of the Octopus Study (coronary artery bypass graft surgery with [n = 73] or without cardiopulmonary bypass [n = 79]). Baseline CRP was dichotomized into a low and a high baseline group, using a cutoff value of 3.0 mg/L.
RESULTS: After coronary artery bypass graft surgery with cardiopulmonary bypass 11 of 53 patients (21%) with low preoperative CRP levels had AF versus 11 of 20 patients (55%) with high baseline CRP levels (p = 0.01). In the off-pump group AF occurred in 4 of 52 patients (8%) who had low baseline CRP levels, versus 8 of 27 patients (30%) with high preoperative CRP levels (p = 0.002). After adjusting for age, the odds ratio (95% confidence interval) was 4.6 (1.4 to 15.3) with cardiopulmonary bypass, 3.7 (0.93 to 14.7) in the off-pump group, and 3.3 (1.4 to 7.6) for both groups together. Continuous baseline CRP was an independent predictor for AF in a multivariate logistic regression model (p = 0.02).
CONCLUSIONS: Patients with high baseline CRP levels are at higher risk of having postoperative AF in both on-pump and off-pump surgery.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
| Dr Lo discloses that he has a financial relationship with Medtronic, Inc.
|
Despite the use of prophylactic antiarrhythmic medication, atrial fibrillation (AF) still occurs in 20% to 40% of patients undergoing coronary artery bypass grafting (CABG). Increased risk of stroke and renal dysfunction, prolonged hospital stay, and accompanying costs are of major concern [1, 2]. Although multiple studies have attempted to determine predictors for the occurrence of postoperative AF, the exact pathogenesis remains unclear [1–4]. In recent years there is growing evidence that AF is associated with an inflammatory state, which can be reflected by plasma C-reactive protein (CRP) levels [5–8]. C-reactive protein levels constitute an independent cardiac risk factor, and studies have shown that they are elevated in patients with postoperative and nonpostoperative AF [6–9]. Postsurgical activation of the complement system after CABG involves CRP. This response is associated with postoperative arrhythmia, and is related to baseline CRP levels [7]. Elevated baseline CRP levels might represent a low-level preoperative inflammatory state that, when altered by the surgical procedure, may lead to a higher incidence of postoperative AF. Baseline CRP levels might thus allow indication of patients with increased risk.
The association between CRP and AF has only been described in on-pump CABG and not in off-pump sur-gery. Therefore, we used data from the Octopus Study to test the hypothesis whether levels of CRP before CABG surgery with and without use of CPB are associated with the occurrence of AF.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
All patients gave written informed consent for the study. The study was approved by the local ethical and research council.
Treatment and Procedures
The goal of surgery was to obtain complete arterial revascularization. With the exception of emergency procedures all operations were performed by cardiac surgeons experienced in both off-pump and on-pump CABG surgery. In the on-pump group, patients received total intravenous anesthesia, including high-dose opioids (sufentanil, 2 to 3 µg/kg at induction, continued with 0.5 µg · kg–1 · h–1 until the end of surgery), whereas in the off-pump group 75% of the patients received thoracic epidural anesthesia combined with low-dose intravenous opioids (sufentanil, 0.25 to 0.5 µg/kg at induction). The CPB patients received 1 mg/kg dexamethasone after induction of anesthesia. Surgical access to the heart was in both groups achieved by midsternal incision.
In the CPB group, the extracorporeal circuit consisted of a membrane oxygenator with integrated heat exchanger, venous reservoir and polyvinyl tubing system, and a nonpulsatile roller pump (Baxter Healthcare Corporation, Irvine, CA) and was primed with crystalloid-colloid mixture. Only the oxygenator and the venous reservoir were heparin coated. Before connection to CPB, porcine heparin (300 IU/kg, Leo Pharmaceutical Products BV, Weesp, The Netherlands) was administered to achieve an activated coagulation time of greater than 450 seconds (Hemochron 400, International Technidyne Corp, NJ). Cardiopulmonary bypass was managed according to the 
-stat principle, with a minimal nasopharyngeal temperature of 32°C; CPB flow rate was kept at 2.4 L · min–1 · m–2. Myocardial protection was achieved with cold (4°C) potassium cardioplegia (Plegisol, Abbott Inc, Chicago, IL). After completion of all distal anastomoses, the aortic cross-clamp was removed, and the proximal anastomoses were performed with a partial occluding clamp. Meanwhile, the patient was rewarmed to 37°C. Heparin was neutralized by 3 mg/kg protamine chloride (Hoffman/La Roche BV, Mijdrecht, The Netherlands). To reduce blood loss, blood was recollected with a suction cardiotomy reservoir in the CPB group, whereas a cell-saving device was used in the off-pump group.
In the off-pump group, after median sternotomy, CABG was performed on the beating, normothermic heart with local cardiac wall immobilization by using the Octopus Tissue Stabilizer (Medtronic, Minneapolis, MN). Before start of the anastomosis, porcine heparin (150 IU/kg) was administered to achieve an activated clotting time of greater than 250 seconds. Heparin was neutralized by 1 mg/kg protamine chloride.
Patient treatment in the intensive care unit was similar for both groups. Patients were monitored with continuous telemetry for at least 48 hours. When detached from electronic monitoring, 12-lead electrocardiograms were made daily for a period of at least 5 postoperative days and in case of clinical suspicion of arrhythmia. An AF episode was defined to last for at least 1 hour and was diagnosed on a physician assessment on the basis of a rhythm strip or an electrocardiogram recording. Anticoagulation was initiated with nadroparin, 2,850 IU subcutaneously, and acetylsalicylic acid, 80 mg, daily. As prophylaxis for postoperative AF all patients were given sotalol, 40 mg twice daily, from the first day until 1 month after surgery.
Measurements
Blood samples were drawn with a syringe from the arterial cannula after induction of anesthesia (baseline). The first 10 mL of the samples was discarded. All samples were immediately cooled on ice and centrifuged twice at 2,000 g for 15 minutes. Plasma was stored at –80°C. Quantitative determination of plasma CRP levels was performed by a high sensitivity CRP assay using a nephelometer (Dade Behring, Inc, Newark, DE). The interassay coefficient of variation is 6.4% [12].
Data Analysis
All analyses were performed using standard computer software (SPSS 9.0, SPSS Inc, Chicago, IL). Normality of the distribution of continuous variables was tested by the Shapiro-Wilk test. All data are presented as mean ± standard deviation or as medians (and interquartile ranges) for nonnormally distributed continuous variables. Because the distribution of CRP levels was highly skewed, logarithmic transformation of CRP was used for the logistic regression. C-reactive protein levels were reported as untransformed values. Differences in continuous and categorized variables were tested by unpaired Student's t test and 
2 test, respectively. In addition to analysis with CRP as a continuous variable, the study cohort was divided into a low (<3.0 mg/L) and high baseline (
3.0 mg/L) CRP group. The cutoff point of 3.0 mg/L was based on the American Heart Association and Centers for Disease Control and Prevention scientific statement suggesting that levels greater than 3 mg/L be considered high [9]. To estimate the risk of developing AF in both groups we used odds ratios and 95% confidence intervals. To account subsequently for possible predictors of postoperative AF (from other studies) and differences found between the high and low baseline CRP groups, multivariate logistic regression analysis was used [1–3, 8]. First the association between postoperative AF and each of the other possible predictors was quantified by univariate analysis. Then, all predictors with p 
0.20 in the univariate analysis were included in the multivariate logistic regression model together with baseline CRP levels.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
After operation, 34 patients had AF, of which 22 were in the CPB group, and 12 in the off-pump group. To determine whether preoperative CRP levels were associated with AF, we subdivided both groups into a low and high baseline (<3.0 or 3.0 mg/L) CRP group.
When both surgical procedures were combined, AF occurred in 15 of 105 patients (14%) in the low baseline CRP group versus 19 of 47 patients (40%) in the high baseline CRP group (crude odds ratio, 4.1 [95% confidence interval, 1.8 to 9.1]; Table 2).
|
Multivariate Logistic Regression
Continuous baseline CRP was a significant predictor of AF in the univariate model. On the basis of known predictors from other studies and differences found between the high and low baseline CRP groups, we then used a multivariate logistic regression to look for possible confounders (Table 3) [1–3, 8]. Only continuous baseline CRP, age, and type of surgery were significant predictors in the model. Hypertension, peripheral vascular disease, use of ß-blockers, hypercholesterolemia, chronic obstructive pulmonary disease, and chronic renal failure (as measured with creatinine levels) did not correlate with the occurrence of AF. The adjusted odds ratio for the high baseline CRP group corrected for age and sex was 3.3 (95% confidence interval, 1.4 to 7.6) for both procedures together. The adjusted odds ratios for high baseline CRP was 4.6 (95% confidence interval, 1.4 to 15.3) with CPB and 3.7 (95% confidence interval, 0.93 to 14.7) in the off-pump group (Table 2).
|
When leaving the hospital, only 4 patients still had AF. These 4 patients all had high baseline CRP levels (1 in the CPB group and 3 in the off-pump group). There was a significant difference in length of hospital stay between patients with and without postoperative AF (median length, 7.0 [interquartile range, 6.0 to 8.0] and 6.0 [interquartile range, 5.0 to 7.0] days, respectively; p < 0.001). However, we could not demonstrate any differences between the low and high baseline CRP groups (median stay, 6.0 [interquartile range, 5.0 to 7.0] days in both groups, p = 0.48).
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Postoperative AF occurs frequently after CABG and represents one of the most common complications of surgical myocardial revascularization. The reported incidences vary between 20% and 40% in patients undergoing CABG, and AF is associated with increased risk of stroke, renal dysfunction, prolonged hospital stay, and increased costs [1, 2]. Prophylactic drug treatment with ß-adrenergic antagonists, sotalol and amiodarone, has been shown to reduce the incidence of postoperative AF; also, nonpharmacologic strategies such as atrial pacing have been developed [13, 14]. Preoperative recognition of patients with a high risk of experiencing postoperative AF could be of considerable value. Limiting (additional) prophylactic antiarrhythmic agents only to a high-risk group could reduce the number to treat reported in the literature, leading to less overtreatment and possible reductions in side effects and costs.
The pathogenesis of postoperative AF appears to be multicausal in origin. Many preoperative and postoperative factors have been suggested to influence its occurrence, but only advanced age has been shown to be a consistent predictor for AF. Other demographic and clinical risk factors such as male sex, hypertension, history of AF, chronic obstructive pulmonary disease, valve surgery, chronic renal failure, and history of heart failure have also been reported [1–4]. The present study also identifies advanced age as a predictor of AF, next to baseline CRP and type of surgery (Table 3). Some studies suggest that an identifiable electrophysiologic substrate, eg, prolonged atrial activation or dispersion of atrial refractoriness, is present in patients who are at highest risk of postoperative AF [15, 16]. Perioperative factors, such as inadequate myocardial protection, electrolyte imbalance, ß-blocker withdrawal, or change in autonomic tone, may precipitate AF in these patients. Recently, Amar and colleagues [17] demonstrated that patients who experienced postoperative AF had different postoperative heart rate variability than patients without postoperative AF, indicating that autonomic influences play a role in the initiation of AF.
There is increasing evidence that AF is associated with an inflammatory state [5–8, 18]. Serum CRP levels can reflect this inflammatory state, and high levels have been shown to constitute an independent cardiac risk factor [9, 18, 19]. In addition to the evidence that elevated baseline CRP levels in healthy persons may lead to cardiac complications, the course after myocardial infarction also may be more complicated [20]. Furthermore, several studies demonstrated a relation between the changes in inflammatory response and the occurrence of AF during cardiac surgery [5, 7]. Gaudino and colleagues demonstrated that the –174G/C interleukin-6 promoter gene variant appears to modulate the inflammatory response to cardiac surgery with CPB and that it influences the development of postoperative AF [5]. The association between administering nonsteroidal antiinflammatory drugs after CABG and a reduced risk of having AF further supports the theory that inflammation may contribute to its pathogenesis [4].
In our previous study, we demonstrated that activation of the classic complement pathway after CABG surgery involved CRP and that changes in levels of CRP, complement, and complement–CRP complexes during the first days after CABG surgery corresponded to postoperative arrhythmia [7]. In addition, CRP levels on the second postoperative day correlated to baseline CRP levels, suggesting that one could distinguish high-CRP responders from low-CRP responders by baseline CRP levels. Our present study demonstrates that (continuous) baseline CRP is an independent predictor for the development of postoperative AF in a multivariate logistic regression model and that patients with high baseline CRP levels have a greater than 3 times higher risk of developing AF, even after correction for age.
The precise mechanism by which inflammation contributes to the development of AF remains unclear. In patients with high baseline CRP levels, more postoperative CRP molecules may localize in atrial tissue, where they promote local complement activation and thus tissue damage, which can lead to fluctuations in the membrane potential [21]. The ligand for CRP remains unknown. In the presence of calcium ions, CRP specifically binds to phosphatidylcholine, in particular when some lysophosphatidylcholine is present [22, 23]. Hence, a supposed ligand for CRP in inflamed tissues may be the membranes of flip-flopped cells or microvesicles derived from these membranes by hydrolyzation by secretory phospholipase A2 [24]. Long-chain acylcarnitines and lysophosphatidylcholines are generated from phosphatidylcholine by phospholipase A2 enzymes, and both can contribute to membrane dysfunction by inhibiting the exchange of sodium and calcium ions in sarcolemmal vesicles and thus lead to the development of arrhythmia [25]. This may explain the association between elevated levels of CRP and the occurrence of AF.
With this study, we are also the first to demonstrate that the correlation between CRP levels and postoperative AF is not CPB dependent, as it was present in both the on-pump and the off-pump groups. In off-pump surgery, it is assumed that the expected reduced trauma, ischemia, and inflammation may significantly reduce the incidence of postoperative AF [26]. Although some studies show a marked reduction in postoperative AF with off-pump surgery, others show comparable incidences with conventional CABG [27]. Conflicting data could be the result of the inhomogeneous mixture of patients in different states of inflammation scheduled for cardiac surgery [28]. In our study the incidence of AF was lower in the off-pump group than in the CPB group, but in the larger patient set of the entire Octopus Study, incidences of AF were the same (on-pump 29 of 139 [21%] and off-pump 28 of 142 [20%], p = 0.79) [10]. There were no additional selection criteria for our study, and patients were taken consecutively from the larger trial. Analysis of the baseline characteristics of both studies did not demonstrate differences that could explain this adequately, so we can only speculate that differences in the preoperative microinflammatory state could play a role.
As the Octopus trial was designed as a pragmatic study, it compared off-pump surgery with conventional CABG, both of which were performed according to what was considered common practice in our institution at that time. This included other factors associated with the on-pump surgery, such as the use of dexamethasone, return of salvaged blood, and the use of epidural analgesia in the off-pump group, which may affect the relationship between inflammation and AF. Therefore, we decided only to study baseline CRP levels because these levels are not affected by these variables. Our study clearly demonstrates a relationship between baseline CRP and the incidence of AF in both types of surgery. Whether the preoperative microinflammatory state, as reflected with CRP levels, leads to a higher postoperative response was demonstrated before and was not a subject of the present study [7].
The use of steroids in the CPB group might be a confounder for the relationship between postoperative inflammation and AF, but it deserves future interest. Recently, Dernellis and Panaretou [18] demonstrated in a randomized trial in a nonsurgical setting that a 4-month treatment with methylprednisolone lowers CRP levels in patients who had experienced persistent AF, thereby successfully reducing the incidence of recurrent and permanent AF. In our study, a single bolus of dexamethasone was given in the CPB group, but the incidence of AF remained higher than in the off-pump group. Apparently, a single bolus of steroids may only delay or have a limited effect on the postoperative inflammatory response, not altering the incidence of AF [29]. Alternatively, it may be suggested that not the magnitude of the postoperative inflammatory response but the global modification of the preoperative microinflammatory state is responsible for an altered incidence of AF. However, further studies are needed to confirm these hypotheses.
In conclusion, we were able to demonstrate that patients with high baseline CRP levels have an increased risk of having postoperative AF after CABG and this accounts for both on-pump and off-pump procedures.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  Y. S. Choi, J. K. Shim, S. W. Hong, D. H. Kim, J. C. Kim, and Y. L. Kwak Risk factors of atrial fibrillation following off-pump coronary artery bypass graft surgery: predictive value of C-reactive protein and transfusion requirement Eur. J. Cardiothorac. Surg., November 1, 2009; 36(5): 838 - 843. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. E. Tisdale, H. A. Wroblewski, D. S. Wall, K. M. Rieger, Z. T. Hammoud, J. V. Young, and K. A. Kesler A randomized trial evaluating amiodarone for prevention of atrial fibrillation after pulmonary resection. Ann. Thorac. Surg., September 1, 2009; 88(3): 886 - 893. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Anselmi, G. Possati, and M. Gaudino Postoperative inflammatory reaction and atrial fibrillation: simple correlation or causation? Ann. Thorac. Surg., July 1, 2009; 88(1): 326 - 333. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Girerd, P. Pibarot, D. Fournier, P. Daleau, P. Voisine, G. O'Hara, J.-P. Despres, and P. Mathieu Middle-aged men with increased waist circumference and elevated C-reactive protein level are at higher risk for postoperative atrial fibrillation following coronary artery bypass grafting surgery Eur. Heart J., May 2, 2009; 30(10): 1270 - 1278. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Sezai, M. Hata, T. Niino, Y. Kasamaki, T. Nakai, A. Hirayama, and K. Minami Study of the factors related to atrial fibrillation after coronary artery bypass grafting: A search for a marker to predict the occurrence of atrial fibrillation before surgical intervention J. Thorac. Cardiovasc. Surg., April 1, 2009; 137(4): 895 - 900. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Pretorius, B. S. Donahue, C. Yu, J. P. Greelish, D. M. Roden, and N. J. Brown Plasminogen Activator Inhibitor-1 as a Predictor of Postoperative Atrial Fibrillation After Cardiopulmonary Bypass Circulation, September 11, 2007; 116(11_suppl): I-1 - I-7. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. B. Neuman, H. L. Bloom, I. Shukrullah, L. A. Darrow, D. Kleinbaum, D. P. Jones, and S. C. Dudley Jr Oxidative Stress Markers Are Associated with Persistent Atrial Fibrillation Clin. Chem., September 1, 2007; 53(9): 1652 - 1657. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. J. Ahlsson, L. Bodin, O. H. Lundblad, and A. G. Englund Postoperative Atrial Fibrillation is Not Correlated to C-Reactive Protein Ann. Thorac. Surg., April 1, 2007; 83(4): 1332 - 1337. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. G. Koch, L. Li, D. R. Van Wagoner, A. I. Duncan, A. M. Gillinov, and E. H. Blackstone Red Cell Transfusion is Associated With an Increased Risk for Postoperative Atrial Fibrillation Ann. Thorac. Surg., November 1, 2006; 82(5): 1747 - 1756. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. W. Hogue Jr, C. A. Palin, R. Kailasam, J. S. Lawton, A. Nassief, V. G. Davila-Roman, B. Thomas, and R. Damiano C-reactive protein levels and atrial fibrillation after cardiac surgery in women. Ann. Thorac. Surg., July 1, 2006; 82(1): 97 - 102. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Shiroshita-Takeshita, B. J.J.M. Brundel, J. Lavoie, and S. Nattel Prednisone prevents atrial fibrillation promotion by atrial tachycardia remodeling in dogs Cardiovasc Res, March 1, 2006; 69(4): 865 - 875. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. V. Podgoreanu and D. A. Schwinn New Paradigms in Cardiovascular Medicine: Emerging Technologies and Practices: Perioperative Genomics J. Am. Coll. Cardiol., December 6, 2005; 46(11): 1965 - 1977. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
