Ann Thorac Surg 2005;80:1732-1737
© 2005 The Society of Thoracic Surgeons
Original article: Cardiovascular
Changing Incidence, Type, and Natural History of Conduction Defects After Coronary Artery Bypass Grafting
David J. Cook, MD
*
,
Joseph M. Bailon, BSN, CRNA,
Tonia T. Douglas, BSN, CRNA,
Kathleen D. Henke, BSN, CRNA,
John R. Westberg, BScN, CRNA,
Mary E. Shirk-Marienau, MS, CRNA,
Thoralf M. Sundt, MD
Mayo Clinic College of Medicine, Rochester, Minnesota
Accepted for publication April 22, 2005.
* Address correspondence to Dr Cook, Mayo Clinic, 200 1st Street SW, Rochester, MN 55905 (Email: cook.david{at}mayo.edu).
 |
Abstract
|
|---|
BACKGROUND: Cardiac conduction defects occur after cardiac surgery. We hypothesized that population aging and increased use of ß-blockers would increase the incidence of new conduction defects after coronary surgery.
METHODS: We examined the medical records of 800 coronary artery bypass grafting (CABG) patients (400 from 1991 and 400 from 2001). Exclusion criteria included the following: preexisting conduction defect, permanent pacemaker, and perioperative atrial fibrillation, leaving 303 and 269 patients, respectively, included in the two study years. The incidence, type, and persistence of new conduction defects were determined from the preoperative, postoperative, and the predischarge electrocardiogram. Multivariate analysis identified predictors of new defects.
RESULTS: Study populations were well-matched. There was a marked decrease in the incidence of new postoperative conduction defects from 1991 (19%) to 2001 (6%). There was also a change in the most frequently occurring block, from a right-bundle-branch-block in 1991 (10%) to first-degree atrioventricular block (3%) in 2001. Finally, conduction defects in 1991 were more transient. While 19% of 1991 patients showed a conduction defect early postoperatively, only 9% were persistent. In 2001, the incidence of conduction defects at discharge (7%), was equivalent to that early postoperatively (6%). Predictors of new conduction defects included year of operation, age, intraaortic balloon counterpulsation, number of vessels bypassed, and crystalloid cardioplegia.
CONCLUSIONS: Our results were the opposite of those predicted. Our report identifies a changing incidence, type, and natural history of conduction defects after CABG. Our comparison demonstrated a decrease in the incidence of new conduction defects, as well as a qualitative change in the defects identified. Multivariate analysis provided predictors of new conduction defects after CABG.
|
Introduction
|
|---|
Defects in cardiac conduction have been a concern since the early days of cardiac surgery and the need for postcardiac surgical pacing was directly responsible for the development of the first pacemakers. Although complete heart block is relatively rare today, a variety of conduction defects are still common after cardiac surgery [1–6]. These conduction defects are a result of patient factors such as chronic degenerative disease of the heart [7, 8], myocardial ischemia or inadequate myocardial protection [6], direct surgical damage to the conduction system [9–11], and possibly the use of medications such as ß-blockers, calcium channel blockers, and antiarrhythmics which may alter cardiac conduction [12].
Because of their frequency and potential severity, postcardiac surgical conduction defects have been well-described. The majority of those investigations were done during the 1970s and 1980s with little work in the last 5 years. In prior investigations, the reported incidence of postoperative conduction defects has been 4% to 58% [2, 13, 14]. Studies of coronary artery bypass grafting (CABG) patients with at least two hundred and fifty subjects have shown incidence rates of 16% to 45% [3, 5, 6, 15]. The most common new onset postoperative conduction defect described has been right bundle branch block (RBBB) [1, 6, 13]. In a more recent (1998) investigation [16], the incidence of new conduction defects in CABG patients was 36% (65 of 181 patients).
We retrospectively examined the individual medical records of two large cohorts of CABG patients, spaced ten years apart, to identify the incidence of new conduction defects in the perioperative period. A multivariate analysis was performed to identify predictors of new onset perioperative conduction defects. We hypothesized that given the aging of the surgical population [17, 18], and the increased use of beta-blockade, that the incidence of new-onset conduction defects would show an increase between 1991 and 2001.
|
Patients and Methods
|
|---|
After receiving approval from the Mayo Foundation Institutional Review Board, we reviewed the medical records of 800 patients who underwent CABG surgery at St Mary's Hospital, Rochester, Minnesota. The samples consisted of 400 consecutive patients in 1991 and 400 consecutive patients in 2001. We included all adult patients 18 years old or greater, presenting to St Mary's Hospital for CABG surgery during the second, third, and fourth quarters of 1991 and all four quarters of 2001. All procedures were done with the use of cardiopulmonary bypass. To decrease the potential variability introduced with other cardiac procedures, we restricted our study to CABG surgery only. Patients were excluded if they had a preexisting conduction defect, a permanent pacemaker, atrial fibrillation, or developed atrial fibrillation in the perioperative period. If there were missing data points (ie, no postoperative electrocardiogram [ECG]), the data point for that record was not included in the final analysis of that characteristic.
We developed a data collection tool to gather data from the medical records of all included patients. Investigators initially abstracted a group of the same charts to ensure inter-rater reliability. Data were obtained from the patient's paper and electronic records, including the preoperative histories, lab values, ECG reports, surgical reports, anesthesia records, and discharge summaries.
To determine the incidence and persistence of a new cardiac conduction deficit, three ECGs were examined: the preoperative ECG, the first ECG taken after admission to the cardiac surgical intensive care unit, and the last ECG recorded before hospital discharge. If the initial postoperative ECG demonstrated a paced rhythm, then the earliest ECG showing no pacemaker activity was recorded as the immediate postoperative ECG.
Electrocardiograms at the Mayo Clinic are recorded electronically using General Electric-Marquette Medical Systems equipment and the Marquette 12SL ECG analysis program. All ECGs were analyzed by computer using standardized criteria [3, 19] and transferred by phone line to Mayo's core ECG laboratory, where the computer analyses were reviewed by technicians. If any new pathology was demonstrated, or there was any ambiguity in interpretation, they were then reviewed by a cardiologist. Standard American Heart Association definitions for cardiac conduction defects were used.
The preoperative, intraoperative, and postoperative characteristics of the 1991 and 2001 cohorts were compared using a Fisher's exact test for categorical variables and a rank sum test for continuous variables. Logistic regression was used to assess preoperative characteristics and operative factors as potential predictors of a new conduction defect. The dependent variable of interest was any new conduction deficit during either the first postoperative or discharge ECG. Patients from both study years were grouped and preoperative medications, preoperative comorbidities, preoperative lab values, and operative factors were considered as potential predictors in this analysis. For each characteristic, a logistic regression model was used to assess the univariate association of the given characteristic with new onset conduction defect, after adjusting for the year of the procedure and the procedure year-by-characteristic interaction. Multiple logistic regression, with backward elimination of nonsignificant variables, was then used to identify a set of independent predictors of a new conduction defect or defects after CABG. To reduce the number of characteristics included in the multivariate analysis, only those found to have some evidence (p < 0.2) of a univariate association with new onset conduction defect were included in the multiple logistic regression model. The analysis was restricted to those patients that did not expire intraoperatively. Out of 800 patient records, a total of 569 patients met inclusion criteria and were included in the multivariate model. In all cases, p values 0.05 or less were considered statistically significant.
|
Results
|
|---|
There were 303 patients from 1991 and 269 patients from 2001 who met the inclusion criteria and were included in the final analysis. Patient demographics and preoperative characteristics are provided in Table 1. Only one patient from 1991 had an incomplete record and was missing preoperative medications and comorbidities. Between 1991 and 2001 patients did not differ with respect to age, gender, and major preexisting conditions, except for hypertension and prior CABG. In 1991 fewer patients were on beta-blockers, and more were on calcium-channel blockers preoperatively than in 2001 (Table 1).
Table 2
describes intraoperative events in the two study populations. Of the 572 patients described, data are complete except that two patients from 1991 are missing American Society of Anesthesiologists status, and one patient from 2001 is missing cardiopulmonary bypass (CPB) time. Between 1991 and 2001 the mean number of grafts performed during CABG was equivalent between groups although the CPB time was decreased in 2001. While total CPB time was shorter in 2001, the aortic cross-clamp time was significantly longer (50.1 vs 45.7 minutes, respectively). Finally, between 1991 and 2001 there was a shift in the type of cardioplegia used with 42% of patients having blood cardioplegia in 1991 versus 83% in 2001 (Table 2).
Early, new, postoperative conduction defects are presented in Table 3. It is evident that there has been a significant decrease in the incidence of early, new onset, post-CABG conduction defects from 1991 to 2001, decreasing from 19% to 6% of patients. There was also a change in the most frequently occurring block from a RBBB (10% of patients in 1991) to a first-degree atrioventricular (AV) block in 2001 (3% of patients). At the time of the first postoperative ECG, less than 10% of the patients in either group were undergoing temporary pacing and there was no statistically significant difference between 1991 and 2001.
Table 4
presents data obtained at discharge. At the time of discharge, 9% of patients having CABG in 1991 and 7% of CABG patients in 2001 had a new conduction defect. The most commonly occurring block for the discharge ECG in 1991 was RBBB, which occurred in 6% of patients. In 2001 the most common conduction defects at discharge were first-degree AV block and RBBB, both occurring in 2% of patients. The type of defect seen at discharge that differed most between 1991 and 2001 was RBBB, changing from 6% to 2%; however, this approached, but did not reach, statistical significance (p = 0.053).
Comparison of the first postoperative ECG and the discharge ECG demonstrates that patients operated on in 1991 often had transient conduction defects. Nineteen percent of patients showed a conduction defect in the first postoperative ECG versus 9% at discharge. This effect was not seen in 2001, when the incidence of conduction defects at discharge (7%) was equivalent to the incidence at the first postoperative ECG (6%). A total of 14 patients from both cohorts were discharged with a new conduction defect that was not present in the first postoperative ECG. However, the frequency of this occurrence was too low to allow analysis of potential causes of the discharge defects not seen in the first postoperative ECG.
Table 4 presents medications prescribed at discharge that potentially affected cardiac conduction in the 1991 and 2001 patient populations. Between 1991 and 2001 there was approximately a three-fold increase in the use of beta-blockers, while approximately twice as many patients in 2001 were discharged on calcium channel blockers relative to 1991. Additionally, less than half as many patients in 2001 were discharged on digitalis than in 1991. In the study population the use of procainamide at discharge decreased from 7% to 0%, while 12% of patients were discharged on amiodarone in 2001 versus less than 1% in 1991.
Univariate and multivariate analyses of potential predictors of a new conduction defect after CABG is provided in Tables 5 and 6. 
All patients and all new conduction defects that occurred were included in the analysis. Twenty-five variables were entered into the univariate analysis; of those, five met the statistical criteria for further analysis. Those were year of surgery, age, cardiopulmonary bypass time, cardioplegia used, need for intraaortic balloon counterpulsation, and number of vessels bypassed. The multivariate analysis demonstrated that year of operation was an independent determinant of the number of new conduction defects seen after CABG surgery (p = 0.034) (Table 6). Other independent predictors of new, post-CABG conduction defects were increased age (p = 0.002), crystalloid cardioplegia (p = 0.046), the requirement for a postoperative intraaortic balloon pump (IABP) (p = 0.026), and an increased number of coronary vessels bypassed (p = 0.046). In a separate analysis looking only at new conduction defects seen in the immediate postoperative period, it was also found that the absence of beta-blockade was a predictor of a new post-CABG conduction defect.
|
Comment
|
|---|
This study has demonstrated a marked decrease in the incidence of new onset conduction defects in CABG patients between 1991 and 2001. Early postoperatively, the incidence of new conduction defects in 2001 was approximately one third that of 1991 (18.6% vs 6.3%, respectively). Additionally, at discharge, 7% showed a conduction deficit in 2001 versus 9% in 1991.
Interestingly, our study findings were the opposite of what we predicted. We hypothesized that the incidence of new conduction defects would increase over the 10-year period, secondary to the aging of the surgical population and the greater use of perioperative beta-blockade. While it is generally true that the surgical population is aging [4, 17, 18] this was not demonstrated in our study population. This deviation from national trends should not be a function of patient selection because consecutive cases were reviewed. That said, older patients more frequently have preexisting conduction defects and preoperative or postoperative atrial fibrillation [2], and because these conditions were exclusion criteria we may have excluded more older patients in the study year 2001, making the cohort ages more uniform. Nevertheless it remains true that advanced age is a strong predictor of new conduction defects after CABG.
In addition to the marked decrease in the incidence of new conduction deficits, there was also a qualitative change in the most commonly occurring defect over the 10-year period of the investigation. As has been previously reported [1, 3, 6, 13], we found that RBBB was the most frequent new postoperative onset conduction defect in 1991. While in 2001 the most common new conduction defect was first-degree atrioventricular block.
The third major observation was that in 1991 most postoperative conduction defects were transient in nature. In the 1991 cohort, at the time of the first, nonpaced postoperative ECG, 19% of patients showed a new conduction defect; however, only 9% of patients were discharged home with a defect present. This effect was not seen in 2001. While 6% of patients showed a defect in their first postoperative ECG, 7% of CABG patients from 2001 had a defect at discharge, demonstrating a less transient nature to the conduction defect in 2001 patients.
This study cannot identify the mechanism for the change in the incidence, the changing character of the conduction disturbance, or the reason for the more transient nature of the problem in 1991 relative to 2001. However, the multivariate analysis' identification of predictors is suggestive, as is the review in the change in medication profile over the 10-year period.
Based on the odds-ratio, the intraoperative need for IABP use corresponded to the greatest likelihood of the development of a new onset conduction defect. This is presumably because the need for an IABP is an expression of significant cardiac disease or failure. Because no study patients had conduction defects preoperatively, this presumably reflects perioperative cardiac injury. While IABP use was the most powerful predictor, it is an infrequent event. Other predictors of new onset conduction defects were far more common, but less powerful, predictors. Those were, in decreasing strength of correlation, year of operation, crystalloid cardioplegia, increased patient age, and number of vessels bypassed.
Because the ages of the two study populations were equivalent, and because age was an equal predictor of conduction defects in both 1991 and 2001, age probably does not contribute to the differences in results between 1991 and 2001. The two study years also did not differ with regard to any other independent determinant of new conduction defects except for cardioplegia management. Crystalloid cardioplegia has been previously associated with right bundle branch block [2, 10]. While the study size is not powered to demonstrate a link between cardioplegia and a specific conduction defect, the dramatic shift away from crystalloid cardioplegia from 1991 to 2001 together with the marked decrease in right bundle branch blocks, strongly suggests that the crystalloid cardioplegia might have been responsible for the much higher incidence of new conduction defects seen in 1991.
This study also demonstrates that medication regimens for CABG patients have changed markedly over the 10-year study period. From 1991 to 2001, the percentage of patients discharged on beta-blockers has tripled, and twice as many patients were discharged on calcium channel blockers. While the nature of the study design cannot demonstrate that the increased use of beta-blockers in 2001 was specifically responsible for the persistent first-degree AV block in that cohort, this effect would be consistent with the mechanism of action and adverse effects of beta-blockade [20], which slows sinus nodal discharge and can delay intraatrial and AV-nodal conduction. Further prospective, clinical investigation would be required to demonstrate this conclusively.
Certainly there are limitations of a database study such as this one. First, database studies can never prove causation, only a strength of association. Second, we examined variables which we thought might contribute to new conduction defects but we did not examine many others. In hindsight, it might have been useful to have acquired more data on cardioplegia; data such as cardioplegia temperature and whether it was given retrograde as well as anterograde. Third, we chose to exclude any patient with preoperative or perioperative atrial fibrillation, or preoperative conduction defects, so our study population may not be completely representative of the current CABG surgical population and the incidence might be higher in a general population. Finally, the study was not designed or powered to assess outcome variables such as length of stay, postoperative myocardial function, or longer term need for pacing. In spite of those limitations, this investigation makes three significant observations. First, the incidence of new onset conduction defects in CABG patients has decreased markedly from 1991 to 2001. Second, qualitatively, the nature of the conduction deficit has changed. Third, the conduction defects demonstrated in 1991 were more transient than those demonstrated in 2001. Current medical-surgical management in CABG patients has decreased the possibility of developing a postoperative conduction defect. While the frequency of new onset conduction defects has significantly decreased, and most positive predictors are fixed, the predictors revealed by this study provide information on those patients most susceptible to new onset conduction defects.
|
Notice From the American Board of Thoracic Surgery
|
|---|
The 2005 Part I (written) examination will be held on Monday, December 5, 2005. It is planned that the examination will be given at multiple sites throughout the United States using an electronic format. The closing date for registration was August 1, 2005. Those wishing to be considered for examination must apply online at www.abts.org.
To be admissible to the Part II (oral) examination, a candidate must have successfully completed the Part I (written) examination.
A candidate applying for admission to the certifying examination must fulfill all the requirements of the Board in force at the time the application is received.
Please address all communications to the American Board of Thoracic Surgery, 6333 N St. Clair St, Suite 2320, Chicago, IL 60611; telephone: (312) 202-5900; fax: (312) 202-5960; e-mail: mailto:info{at}abts.org.
|
Acknowledgments
|
|---|
Financial support was obtained from the Department of Anesthesia, Mayo Clinic Foundation.
|
References
|
|---|
- Baerman JM, Kirsh MM, de Buitleir M, et al. Natural history and determinants of conduction defects following coronary artery bypass surgery Ann Thorac Surg 1987;44:150-153.[Abstract]
- Pires LA, Wagshal AB, Lancey R, Huang SK. Arrhythmias and conduction disturbances after coronary artery bypass graft surgeryepidemiology, management, and prognosis. Am Heart J 1995;129:799-808.[Medline]
- Chu A, Califf RM, Pryor DB, et al. Prognostic effect of bundle branch block related to coronary artery bypass grafting Am J Cardiol 1987;59:798-803.[Medline]
- Emlein G, Huang SK, Pires LA, Rofino K, Okike ON, Vander Salm TJ. Prolonged bradyarrhythmias after isolated coronary artery bypass graft surgery Am Heart J 1993;126:1084-1090.[Medline]
- Caspi J, Amar R, Elami A, Safadi T, Merin G. Frequency and significance of complete atrioventricular block after coronary artery bypass grafting Am J Cardiol 1989;63:526-529.[Medline]
- Caspi Y, Safadi T, Ammar R, Elamy A, Fishman NH, Merin G. The significance of bundle branch block in the immediate postoperative electrocardiograms of patients undergoing coronary artery bypass J Thorac Cardiovasc Surg 1987;93:442-446.[Abstract]
- Falk RH. Etiology and complications of atrial fibrillationinsights from pathology studies. Am J Cardiol 1998;82:10N-17N.[Medline]
- Mackstaller LL, Alpert JS. Atrial fibrillationa review of mechanism, etiology, and therapy. Clin Cardiol 1997;20:640-650.[Medline]
- Thompson R, Mitchell A, Ahmed M, Towers M, Yacoub M. Conduction defects in aortic valve disease Am Heart J 1979;98:3-10.[Medline]
- Bateman TM, Weiss MH, Czer LS, et al. Fascicular conduction disturbances and ischemic heart diseaseadverse prognosis despite coronary revascularization. J Am Coll Cardiol 1985;5:632-639.[Abstract]
- Satinsky JD, Collins Jr JJ, Dalen JE. Conduction defects after cardiac surgery Circulation 1974;50:II170-II174.
- Slogoff S, Keats AS, Ott E. Preoperative propranolol therapy and aortocoronary bypass operation JAMA 1978;240:1487-1490.[Abstract/Free Full Text]
- Wexelman W, Lichstein E, Cunningham JN, Hollander G, Greengart A, Shani J. Etiology and clinical significance of new fascicular conduction defects following coronary bypass surgery Am Heart J 1986;111:923-927.[Medline]
- Zeldis SM, Morganroth J, Horowitz LN, et al. Fascicular conduction disturbances after coronary bypass surgery Am J Cardiol 1978;41:860-864.[Medline]
- Gundry SR, Sequeira A, Coughlin TR, McLaughlin JS. Postoperative conduction disturbancesa comparison of blood and crystalloid cardioplegia. Ann Thorac Surg 1989;47:384-390.[Abstract]
- Mustonen P, Hippelainen M, Vanninen E, Rehnberg S, Tenhunen-Eskelinen M, Hartikainen J. Significance of coronary artery bypass grafting-associated conduction defects Am J Cardiol 1998;81:558-563.[Medline]
- Sakamoto S, Matsubara J, Matsubara T, et al. Coronary artery bypass grafting in octogenarians Cardiovasc Surg 2001;9:487-491.[Medline]
- Ascione R, Rees K, Santo K, et al. Coronary artery bypass grafting in patients over 70 years oldthe influence of age and surgical technique on early and mid-term clinical outcomes. Eur J Cardiothorac Surg 2002;22:124-128.[Abstract/Free Full Text]
- Hecht HH, Kossmann CE, Childers RW, et al. Atrioventricular and intraventricular conduction. Revised nomenclature and concepts Am J Cardiol 1973;31:232-244.[Medline]
- Frishman WH. ß-adrenergic receptor blockersadverse effects and drug interactions. Hypertension 1988;11(suppl II):II-21-II-29.
Top
Abstract
Introduction
