CLINICAL RESEARCH: ELECTROPHYSIOLOGY
Predictors of new malignant ventricular arrhythmias after coronary surgery
A case-control study
Raimondo Ascione, MD, MCh*,*,
Barnaby C. Reeves, PhD*,
Kirkpatrick Santo, FRCS*,
Nouman Khan, FRCS* and
Gianni D. Angelini, MD, MCh, FRCS*
* Bristol Heart Institute, Bristol Royal Infirmary, Bristol, United Kingdom
Manuscript received May 2, 2003;
revised manuscript received October 20, 2003,
accepted November 3, 2003.
* Reprint requests and correspondence: Dr. Raimondo Ascione, Bristol Heart Institute, Bristol Royal Infirmary, Bristol BS2 8HW, United Kingdom.
R.Ascione{at}bristol.ac.uk
 |
Abstract
|
|---|
OBJECTIVES: We sought to investigate the relationship between perioperative factors and the occurrence of ventricular tachycardia (VT) and ventricular fibrillation (VF), as well as the impact of VT/VF on early and late mortality.
BACKGROUND: Both VT and VF are rare but serious complications after coronary artery bypass graft surgery (CABG), and their etiology and implications remain uncertain.
METHODS: Data on 4,411 consecutive patients undergoing CABG (1,154 [25.8%] had off-pump surgery) between April 1996 and September 2001 were extracted from a prospective database and analyzed. Odds ratios (ORs) describing associations between possible risk factors and VT/VF were estimated separately. Factors observed to be significantly associated with VT/VF were further investigated using multivariate logistic regression.
RESULTS: Sixty-nine patients suffered VT/VF (1.6%). There were 61 (1.4%) in-hospital/30-day deaths, 15 among patients who had postoperative VT/VF (21.7%). Patient factors independently associated with an increase in the odds of VT/VF included age <65 years, female gender, body mass index <25 kg/m2, unstable angina, moderate or poor ejection fraction, and the need for inotropes and an intra-aortic balloon pump (OR 1.72 to 4.47, p < 0.05). After adjustment, off-pump surgery was associated with a substantial but nonsignificant protective effect against VT/VF (OR 0.53, 95% confidence interval [CI] 0.25 to 1.13; p = 0.10). Actuarial survival at two years was 98.2% among patients who had VT/VF and who survived to discharge/30 days, compared with 97.0% for the control group (adjusted hazard ratio 0.96 (95% CI 0.40 to 2.31, p = 0.92).
CONCLUSIONS: The incidence of VT/VF is low in patients undergoing coronary surgery but is associated with high in-hospital mortality. The late survival of the discharged VT/VF patients compares favorably with that of controls.
|
Abbreviations and Acronyms
| | BMI | = body mass index | | CABG | = coronary artery bypass graft surgery | | CI | = confidence interval | | EF | = ejection fraction | | IABP | = intra-aortic balloon pump | | ICD | = implantable cardioverter-defibrillator | | LV | = left ventricular | | MI | = myocardial infarction | | OR | = odds ratio | | PATS | = Patient Activity Tracking System | | VF | = ventricular fibrillation | | VT | = ventricular tachycardia |
|
Sustained ventricular tachycardia (VT) and ventricular fibrillation (VF) are life-threatening complications (1,2). Many cardiac diseases may increase the risk of ventricular arrhythmias, but acute ischemia and previous myocardial infarctions (MIs) are by far the most common (1). New ventricular arrhythmias may also be observed after coronary artery bypass graft surgery (CABG), varying from ventricular ectopy (3) to tachyarrhythmias such as nonsustained VT, sustained VT, and VF (1–3). Patient and operative factors, inadequate myocardial protection, metabolic imbalance, residual ischemia, and surgical factors like incomplete revascularization and suboptimal anastomosis are all potential causes of VT/VF during the recovery period after CABG (1–4). In addition to the importance of the clinical consequences of VT/VF, these complications may have an economic effect by prolonging the intensive care unit and hospital stay. Evidence in the literature describing the frequency of new VT/VF in the recovery period after CABG remains anecdotal (5–7), and the lack of information on possible risk factors for VT/VF is alarming in view of the life-threatening nature of this complication.
The aims of this study were two-fold: first, to investigate associations between patient and operative factors and the occurrence of VT/VF; second, to describe the impact of VT/VF on early and late mortality in a large population undergoing myocardial revascularization.
|
Methods
|
|---|
Patient selection.
A nested case-control study was carried out using data collected prospectively from an unselected cohort of patients undergoing CABG. By defining the groups to be compared on the basis of the outcome of interest, this study design can investigate a wide range of factors that may be associated with a rare outcome. The representative nature of the cohort meant that the incidence of VT/VF could also be estimated. All data were collected prospectively at the time of operation and entered into the Patient Activity Tracking System (PATS) database (Dendrite Systems, London, United Kingdom), which is maintained routinely in our institution to document cardiac surgical activity. The analysis included all patients undergoing CABG between April 1, 1996, and September 13, 2001, with no previous history of VT/VF. During the study period, surgeons in our institution began to carry out off-pump CABG and have used this new technique with increasing frequency, as well as in a wider range of patients, as they have gained more experience. Data were included for all CABG patients, irrespective of whether they underwent conventional CABG or off-pump CABG.
Cases were defined as patients who required advanced cardiac life support interventions to terminate the malignant arrhythmia and to stabilize the patient in the postoperative period before discharge. Controls were defined as all other patients in the cohort. The prevalence of preoperative characteristics, surgical factors, and aspects of postoperative management were compared between cases and controls.
Anesthetic and surgical technique.
Anesthetic and surgical techniques were standardized for all patients and have been previously reported (8).
For patients undergoing on-pump CABG, cardiopulmonary bypass was instituted using ascending aortic cannulation and two-stage venous cannulation of the right atrium. Non-pulsatile flow was used, and flow rates throughout bypass were 2.4 l/m2/min. A systemic temperature was kept between 34°C and 36°C. Myocardial protection was achieved by using intermittent anterograde hyperkalemic warm-blood cardioplegia (8). For patients undergoing off-pump CABG, the target vessel was exposed and snared above the anastomotic site using a 4-0 prolene suture with a soft plastic snugger. The coronary artery was then opened; an intracoronary shunt was used in case of relative electrocardiographic (ECG) or hemodynamic instability or excessive bleeding; and the anastomosis was performed using a pressure stabilizer (9).
Postoperative management.
At the end of surgery, patients were transferred to the intensive care unit. Forced air warming was used until a stable nasopharyngeal temperature of 37°C had been reached. Patients were extubated as soon as they met the following criteria: hemodynamic stability, no excessive bleeding (<80 ml/h), normothermia, and consciousness with pain control. Fluid management postoperatively consisted of 5% dextrose infused at 1 ml/kg/h, with additional Gelofusine or blood to maintain normovolemia and a hematocrit level >24%. Potassium and magnesium deficiency was promptly treated, as necessary, to maintain electrolyte balance within the normal range.
Clinical data collection, monitoring, and definitions.
A wide range of potential risk factors for VT/VF (Table 1), including preoperative interventions indicative of low cardiac output (i.e., requirement for intra-aortic balloon pump [IABP] or inotropic support [>5 µg/kg/min of dopamine] for >30 min to maintain systolic blood pressure >90 mm Hg and a cardiac index >2.2 l/min/m2), and intraoperative and postoperative data were entered prospectively into the PATS data base. According to our unit protocol, blood pressure and heart rate and rhythm were continuously monitored and displayed on a screen inclusive of an automated arrhythmia detector (Solar 8000 Patient Monitor, Marquette Medical Systems, Milwaukee, Wisconsin) during the first 72 h postoperatively. Automatic printing of the ECG was related to the predefined alarm levels. Twelve-lead ECG recordings were obtained preoperatively, 2 h postoperatively, and then daily thereafter until discharge.
Each episode of arrhythmia was printed out and interpreted by an intensivist. Sustained VT was defined as a uniform tachyarrhythmia of ventricular origin that had a duration of >20 s, unless terminated earlier because of hemodynamic collapse. When VT/VF occurred, standard advanced cardiac life support was commenced to terminate the arrhythmia and to stabilize the patient. Ventricular tachycardia was managed according to standard protocols of our unit. These protocols aim to achieve assessment and treatment of any acute reversible or correctable cause of arrhythmia, including metabolic alterations, myocardial ischemia/infarction, congestive heart failure, and potentially proarrhythmic medications. Patients who had an isolated episode of in-hospital VT/VF are discharged with antiarrhythmic medications for a period of at least six months, according to the outcome of follow-up in the cardiology outpatient clinic. Patients with more than one episode of VT/VF are referred to a cardiologist for an electrophysiologic study to assess the need for an implantable cardioverter-defibrillator (ICD).
Sample size.
Even with over 4,000 patients (Table 1), the power of the analysis was limited because of the rarity of VT/VF (1.6%). The power of the analyses to detect clinically important associations depended on both the prevalence of the risk factor among controls and the magnitude of the risk conferred, as well as the incidence of VT/VF. The analyses had 80% power to detect an odds ratio (OR) of  3.3 for risk factors with a prevalence of 5% at a 5% (two-tailed) significance level, or an OR of 2.2 for risk factors with a prevalence of 20%. Risk factors with an OR of <2.0 were unlikely to be detected by the analyses.
Statistical analysis.
We chose not to include deaths in a composite outcome measure, because the occurrence of VT/VF is almost always identified clinically before death. Therefore, it would have been inappropriate to pool patients who died two or more days after the operation, having shown no early signs of VT/VF, with patients in whom VT/VF was clinically manifested. Patients who died in the hospital were therefore included in case or control groups on the same basis as patients who survived (i.e., according to whether or not they showed signs in the immediate perioperative and postoperative period of having experienced VT/VF.
More variables are available in the PATS database than are described here. A decision about the key variables of interest to investigate was made before conducting any analyses. All variables selected at the outset of investigation are listed in Table 1. Some continuous variables were grouped for analysis. Age, body mass index (BMI), Parsonnet score, and ejection fraction (EF) were dichotomized (<65 vs. 65 years, BMI <25 vs. 25 kg/m2, Parsonnet score  10 vs. >10, and good vs. moderate or poor EF, respectively) for the multivariate analyses for ease of interpretation and because ORs in the multivariate models for adjacent categories were similar.
All analyses were carried out by logistic regression modeling, adjusting the standard errors of estimates of ORs for clustering of patients within consultant teams. Multivariate analyses considered all predictors that separately showed any evidence of an association with VT/VF (p < 0.20). Predictors were retained in the final "adjusted" model if they were independently associated with the occurrence of VT/VF (p < 0.05). When two or more predictors were strongly correlated, only one factor (the one that was most strongly associated with VT/VF) was retained in the final model.
The risk of postoperative VT/VF for subsequent mortality was then investigated by survival analysis, first, considering all deaths, and second, only the deaths that occurred after hospital discharge. For the latter analysis, the same "time zero" was used (i.e., date of operation), but the analysis was confined to patients who survived 30 days or to hospital discharge, if the discharge occurred more than 30 days after the operation. Univariate and multivariate Cox regression models were fitted, stratified by consultant team, to estimate survival at one and three years and the hazard ratio for the group that had experienced VT/VF.
A post hoc subgroup analysis was carried out among patients who suffered VT/VF, comparing patients who had VT/VF intraoperatively or postoperatively, to ascertain whether the time of occurrence was influenced by patient factors. The latter group was further classified according to the time of postoperative occurrence of VT/VF (i.e., 24 h postoperatively).
All analyses were carried out using STATA version 7 (STATA Corp., Houston, Texas).
|
Results
|
|---|
Description of study population.
During the study period, a total of 4,467 patients underwent CABG, of whom 1,154 (25.8%) had off-pump surgery. Information on whether or not VT/VF had occurred was missing for 56 patients (1.3%; these patients are not considered further), but data on potential risk factors were complete for more than 99% of the remaining 4,411 cases for 23 of the 25 preoperative patient factors investigated. Exceptions were BMI (4,330 of 4,411 [98.2%]) and severity of coronary disease (4,270 of 4,411 [98.5%]). Sixty-nine of the 4,411 patients (1.6%; 95% confidence interval [CI] 1.2% to 2.0%) had new-onset VT/VF, all as an isolated episode. Overall, there were 61 deaths in the hospital or deaths within 30 days of CABG (1.4%), 15 among the 69 patients who had VT/VF (21.7%). These 15 patients died zero to 43 days after the operation. The intraoperative occurrence of VT/VF had a stronger impact on mortality (5 of 12 [41.6%]), as compared with postoperative VT/VF (10 of 57 [17.5%]).
Predictors of VT/VF.
Several preoperative factors were separately associated with a significant increase in the odds of VT/VF (p < 0.05; ORs ranging from 2.21 to 17.3) (Table 2), including female gender, Canadian Cardiovascular Society classification, New York Heart Association functional class, presence of unstable angina, congestive cardiac failure, preoperative use of IABP, preoperative use of inotropes, Parsonnet score, decreasing preoperative EF, and operative priority. An increasing BMI appeared to be protective against VT/VF (OR 0.57, p = 0.05 for BMI 25 and <30 kg/m2 vs. 20 and <25 kg/m2) (Table 2). There was no evidence that atrial fibrillation before the index operation was more or less likely in patients who experienced VT/VF than in patients who did not.
Multiple logistic regression modeling identified eight preoperative factors that were independently associated with the risk of any VT/VF (Table 2): age <65 years, female gender, BMI <25 kg/m2, preoperative unstable angina, IABP required, inotropes required, Parsonnet score >10, and poor or moderate EF. The ORs for these risk factors ranged from 1.72 (moderate or poor EF) to 4.47 (preoperative requirement for IABP).
Operative risk factors (off-pump surgery and number of grafts) were added to the adjusted model containing preoperative predictors. Off-pump surgery was associated with a substantial but nonsignificant protective effect against VT/VF (OR 0.53, 95% CI 0.25 to 1.13; p = 0.10). The number of grafts carried out did not appear to be an important risk factor for VT/VF. The ORs for all preoperative variables were essentially unchanged when operative factors were added to the multivariate model.
We also investigated associations between the occurrence of VT/VF and postoperative in-hospital outcome. Perioperative MI and atrial fibrillation were much more common in patients who had VT/VF than in patients who did not (MI: 14.5% vs. 1.7%; atrial fibrillation: 36.2% vs. 11.5%, both p < 0.0001). These associations were observed irrespective of whether VT/VF occurred in the first 12 hours after the operation or subsequently. Unfortunately, the database did not include more detailed information on the relative timing of the events. Patients who had VT/VF were more likely to have other serious complications as well (e.g., requiring inotropes, hemodialysis, or reintubation or having transient stroke or multiorgan failure). As with perioperative MI and atrial fibrillation, we were unable to identify whether these other complications preceded or followed VT/VF.
Analysis of VT/VF group.
The timing of the malignant arrhythmias was either at anesthetic induction or during the operation in 12 patients (17.5%) and postoperatively in the remaining 57 (85.6%). Five (42%) of 12 patients who experienced intraoperative VT/VF died in the hospital, as compared with 10 (18%) of 55 patients who had VT/VF on the intensive care unit or ward (p = 0.08) (data on timing of VT/VF were missing for 2 patients). Of the 15 in-hospital deaths within the VT/VF group, two died within the first 3 h postoperatively. The remaining 13 patients died between postoperative days 2 and 43. Among the 57 patients with postoperative VT/VF, 27 (47.4%) had the event within the first 24 h.
A subgroup analysis was carried out between patients with VT/VF either intraoperatively or postoperatively to ascertain whether this time of occurrence was influenced by baseline status (i.e., recent MI, poor left ventricular [LV] function, emergency referral). The comparison between the two subgroups did not show any significant associations between VT/VF and baseline recent MI, poor LV function, or emergency referral, with p values of 0.74, 0.42, and 0.11, respectively. Of the 54 VT/VF patients surviving to discharge or 30 days, 29 (53.7%) were discharged while receiving amiodarone, 15 (27.7%) beta-blockers, and 10 (18.5%) both. None of the VT/VF patients underwent ICD placement.
Late clinical outcome in VT/VF patients.
The overall median duration of follow-up was 3.6 years (interquartile range 2.2 to 4.9 years). The duration of follow-up differed for case and control groups (2.8 vs. 3.6 years, respectively), but only because of the deaths in the hospital or within 30 days. Among survivors to discharge/30 days, the median follow-up was 3.6 years in both groups. In total, there were 332 deaths during follow-up (i.e., there were a further 271 deaths in addition to the 61 that occurred in the hospital or within 30 days). Actuarial survival for patients who had in-hospital VT/VF and who survived to discharge/30 days was 76.8% at 2 years, compared with 96.0% for the control group (unadjusted hazard ratio 4.54, 95% CI 3.39 to 6.07; p < 0.0001). Independent preoperative risk factors for mid-term mortality included younger age, diabetes, unstable angina, congestive cardiac failure, poor EF, hypertension, previous cerebrovascular accident, preexisting respiratory comorbidity, requirement for inotropes, and increasing Parsonnet score. After adjusting for these risk factors, the hazard ratio was 3.52 (95% CI 2.13 to 5.81, p < 0.0001).
The Kaplan-Meier curve for this analysis suggested that excess mortality in the VT/VF group occurred only during the immediate postoperative period (Fig. 1). The analysis was therefore rerun including only those patients who survived to discharge and who survived beyond 30 days. The survival estimates are imprecise for this analysis because the total number of patients who experienced VT/VF and survived to discharge and beyond 30 days was small (n = 54), and there were only four deaths during follow-up in this group (vs. 267 deaths after discharge and beyond 30 days in the control group). Survival for patients who had in-hospital VT/VF was 98.2% at two years, compared with 97.0% for the control group (unadjusted hazard ratio 1.13, 95% CI 0.55 to 2.34; p = 0.74). Independent preoperative risk factors for mid-term survival after hospital discharge included unstable angina, congestive cardiac failure, moderate and poor EF, previous cerebrovascular accident, requirement for inotropes, increasing Parsonnet score, and operative priority. After adjusting for these risk factors, the hazard ratio was 0.92 (95% CI 0.43 to 1.99, p = 0.83).
View larger version (25K):
[in this window]
[in a new window]
|
Figure 1 Kaplan-Meier survival estimates: all deaths (dashed line = ventricular tachycardia (VT)/ventricular fibrillation (VF) cases; solid line = controls). CI = confidence interval
|
|
|
Discussion
|
|---|
New occurrences of VT/VF may affect the recovery period after CABG. Despite its life-threatening nature, there are few studies in the literature addressing this topic (5–7). Consequently, associations between patient and operative factors and the occurrence of VT/VF, as well as its impact on early and late mortality, remain uncertain. This is the first large study based on prospectively collected data to investigate the new onset of VT/VF after coronary surgery.
The observed incidence of sustained VT/VF in the present cohort was 1.6%, considerably less than previously reported. Steinberg et al. (10) reported an incidence of 3.1% of in-hospital sustained VT/VF after CABG at a single institution in a population of 382 consecutive patients. A similar or higher incidence has been reported by others (3,5,11). A potential explanation might be the use, at our institution, of intermittent anterograde warm-blood cardioplegia, which is known to reduce the occurrence of ventricular arrhythmias and the need for administration of lidocaine hydrochloride for treatment of ventricular arrhythmias after CABG (12). The use of off-pump coronary surgery in up to 23% of the entire population may also have contributed, as the use of this technique was observed to have a substantial protective effect against VT/VF, although this did not reach significance.
The present study identifies several independent predictors of new-onset, sustained VT/VF in patients undergoing coronary surgery. It is tempting to consider whether these risk factors can help to clarify the clinical profile of high-risk patients, perhaps allowing a preoperative risk stratification method for VT/VF, with the possibility of prescribing a prophylactic regimen for those at highest risk. Unfortunately, the risk factors for VT/VF that we observed do not help very much to identify a subgroup of patients at high risk of VT/VF. Formally, the area under the receiver-operating characteristics curve for the model was 0.78. Maximizing the proportion of patients correctly classified by the model (98.5%) yielded a sensitivity of only 1.5%; on the other hand, setting an acceptable sensitivity (i.e., 70%) yielded a specificity of only 71% (the statistics overestimate the performance of the model since they are calculated from the same dataset on which the model was estimated).
The poor prediction of individual patients arises from the rarity of VT/VF and the fact that the majority of the risk factors were common; age <65 years, female, BMI <25 kg/m2, unstable angina, and moderate or poor EF had prevalences of over 20% in the study population. Patients who needed preoperative inotropic support or an IABP were less frequent, and these risk factors were associated with higher ORs. Therefore, it may be reasonable to regard these patients as being at high risk of VT/VF, but these factors alone have only about half the predictive power than all of the predictors. If it is subsequently shown that postoperative atrial fibrillation usually precedes VT/VF, this would also be an important, albeit postoperative, risk factor for VT/VF.
Our study strongly suggests that patients who had VT/VF were more likely to die in the hospital (21.7%), as compared with patients who did not (1.0%). This finding is consistent with the 25% in-hospital mortality rate reported by Steinberg et al. (10). Of interest is that intraoperative occurrence of VT/VF had a stronger impact on in-hospital mortality than did postoperative VT/VF.
The presence of baseline unstable angina was found to be an independent predictor of the occurrence of VT/VF. The extent of coronary disease, time of cardioplegic arrest, cardioplegic solution, route of delivery, and temperature of the perfusate may all contribute to nonhomogeneous myocardial protection (13), triggering many potential pathophysiologic mechanisms of ventricular arrhythmias in unstable angina patients. Unstable angina-induced ischemia may indeed determine alterations in cellular electrophysiology, local accumulation of catecholamines, abnormal automaticity within surviving Purkinje fibers, or reentry within the ischemic myocardium (14). Steinberg et al. (10) reported that grafting a non-collateralized, totally occluded vessel supplying an infarct zone is independently associated with the occurrence of VT. They hypothesized that resumption of blood flow to areas previously poorly perfused could potentially restore electrophysiologic function to cells embedded within the borders of myocardial scar and create re-entrant circuits.
In contrast to other reports, a history of MI was not associated with occurrence of VT/VF in our study (15,16). This might be due to the fact that most of our pre-infarcted patients were operated on at least one to two weeks after acute MI. Cheema et al. (14) have highlighted the poor prognostic significance of VT occurring in the setting of acute MI (within the first days), as opposed to a negative prognostic outcome, as suggested by Bigger et al. (17), when occurring in the setting of healing MI (more than seven days after MI).
We also found a significant association between moderate to poor LV function and the occurrence of VT/VF. The presence of impaired LV function is not synonymous with irreversible damage, as hibernated or stunned myocardium may also contribute to poor LV function (18). Steinberg et al. (10) suggested the importance of not only coronary perfusion to maintain the electrical viability of critical arrhythmogenic tissue but also a reversible state of "electrical hibernation," which may only be possible in the absence of irreversible myocardial injury. Our findings seem to support this hypothesis. Others, however, have demonstrated the association between revascularization of an area of previous MI and the development of sustained VT/VF (19).
Another interesting finding of the present study was the significant association between female gender and VT/VF after coronary surgery. The gender-specific electrophysiologic structure of the heart and/or hormonal effects on modulation ionic channel function may help to explain differences such as the increased prevalence of symptoms of congenital long QT syndrome in women and the increase in episodes of supraventricular tachycardia in the perimenstrual period (20,21). This seems to be supported also by experimental data. Saba et al. (22) found that in female mice, estrogen prolongs atrioventricular nodal conduction and the right ventricular effective refractory period. In contrast, Aronson and Burger (23) found that complex ventricular ectopy and episodes of ventricular tachycardia are significantly lower in women. However, this was not a surgical series; therefore, patients were not exposed to all of the potential triggering factors that act exclusively as a consequence of surgery.
We identified age <65 years as an independent risk factor for VT/VF and overweight/obesity (25 kg/m2) as a protective factor. These findings are quite interesting, as no similar results are reported in the literature. Limited development of collaterals in younger patients with severe coronary disease might make these patients less tolerant to ischemia, which, in turn, might trigger the occurrence of VT/VF.
Patients with in-hospital VT/VF and discharged home appear to have a late survival comparable to the controls. There are several potential explanations for this finding. First, all of the VT/VF patients were discharged on antiarrhythmic drugs (i.e., amiodarone [53.7%], beta-blockers [27.7%], or both [18.5%]). A recent meta-analysis of 13 randomized trials using amiodarone reported a 13% overall reduction in total mortality and a 29% reduction in antiarrhythmic death (24). Others have highlighted the role of beta-blockers in improving survival in VT/VF patients (25). It has also been suggested, however, that a favorable response to antiarrhythmic therapy simply identifies a lower risk population, and no direct benefit is gained by treatment with antiarrhythmic drugs (26). This could be true in our series, as all patients underwent revascularization, which could have removed the electrical milieu responsible for future arrhythmic events. Furthermore, a postoperative metabolic or electrolyte imbalance was promptly treated with definitive removal of a further potential triggering factor of VT/VF. These results also support our policy of not implanting an ICD in patients with an isolated episode of VT/VF.
Case-control studies are susceptible to bias and confounding (27), and it is important to consider the extent to which these problems may have affected our findings. We think that confounding from a selection bias is an unlikely explanation for the findings observed in this study, because cases and controls were drawn from exactly the same population, an advantage of the "nested" design. However, we cannot rule out the possibility of incomplete adjustment for confounding in the multivariate analyses. The prospective nature of data collection also protects against bias, as the case or control status of participants could not have influenced the measurement of outcome or predictor factors. Bias could also have arisen from missing data, especially with respect to missing data for the occurrence of VT/VF. However, our use of an automated system for detection of arrhythmias makes this very unlikely.
Study limitations.
One of the limitations of the present study is that despite the large size of the cohort, it could only identify factors that conferred a substantial increase in risk because of the rarity of VT/VF. Other factors that were not found to be significantly associated with VT/VF from a statistical perspective may nevertheless confer a clinically important increase in risk (e.g., congestive cardiac failure and MI). These potential predictors of VT/VF need to be investigated further in larger cohorts. Also, we cannot rule out the possibility that we missed late, self-converting, nonsustained episodes of VT/VF in our study. However, this is unlikely for the first 72 h postoperatively due to continuous monitoring with an automated arrhythmia detector. Furthermore, nonsustained VT/VF was not part of the primary outcome of the present study.
Conclusions.
This study shows a relatively low incidence of VT/VF in patients undergoing coronary surgery, but there was a high in-hospital mortality associated with this complication. In contrast, the late survival of discharged VT/VF patients compares favorably with that of controls. Several factors were found to be independently associated with the occurrence of VT/VF. Knowledge of important risk factors for VT/VF may be useful in order to optimize a patient's preoperative physiologic status, the surgical techniques used, and perioperative and postoperative management.
|
Footnotes
|
|---|
The Garfield Weston Trust and the British Heart Foundation supported this work.
|
References
|
|---|
1. Welch PJ, Page RL, Hamdan MH. Management of ventricular arrhythmias. J Am Coll Cardiol. 1999;34:621–630[Abstract/Free Full Text]
2. O'Rourke RA. Role of myocardial revascularization in sudden cardiac death. Circulation. 1992;85(Suppl I):I112–I117
3. Rubin DA, Nieminski KE, Monteferrante JC. Ventricular arrhythmias after coronary artery bypass surgery: Incidence, risk factors, and long-term prognosis. J Am Coll Cardiol. 1985;6:307–310[Abstract]
4. Costeas XF, Schoenfeld MH. Usefulness of electrophysiologic studies for new-onset sustained ventricular tachyarrhythmias shortly after coronary artery bypass grafting. Am J Cardiol. 1993;72:1291–1294[CrossRef][Medline]
5. Ducceschi V, D'Andrea A, Liccardo B, et al. Ventricular tachyarrhythmias following coronary surgery. Int J Cardiol. 2000;73:43–48[CrossRef][Medline]
6. Ferraris VA, Ferraris SP, Gilliam HS, Berry WR. Predictors of postoperative ventricular dysrhythmias: A multivariate study. J Cardiovasc Surg. 1992;32:12–20
7. Kron IL, DiMarco JP, Harman KP. Unanticipated postoperative ventricular tachyarrhythmias. Ann Thorac Surg. 1984;38:317–322[Abstract]
8. Ascione R, Caputo M, Calori G, et al. Predictors of atrial fibrillation after conventional and beating heart coronary surgery: A prospective, randomized study. Circulation. 2000;102:1530–1535[Abstract/Free Full Text]
9. Watters MPR, Ascione R, Ryder IG, et al. Hemodynamic changes during beating heart coronary surgery with the ‘Bristol technique. ’ Eur J Cardiothorac Surg. 2001;19:34–40
10. Steinberg JS, Gaur A, Sciacca R, Tan E. New-onset sustained ventricular tachycardia after cardiac surgery. Circulation. 1999;99:903–908[Abstract/Free Full Text]
11. Kaul TK, Fields BL, Riggins LS, et al. Ventricular arrhythmia following successful myocardial revascularization: Incidence, predictors, and prevention. Eur J Cardiothorac Surg. 1998;13:629–636
12. Calafiore AM, Teodori G, Mezzetti A, et al. Intermittent antegrade warm blood cardioplegia. Ann Thorac Surg. 1995;59:398–402[Abstract/Free Full Text]
13. Guiraudon GM. Coronary artery bypass grafting in the management of ventricular arrhythmias in patients with coronary artery disease. Curr Opin Cardiol. 1992;7:945–949[Medline]
14. Cheema AN, Sheu K, Parker M, et al. Nonsustained ventricular tachycardia in the setting of acute myocardial infarction: Tachycardia characteristics and their prognostic implications. Circulation. 1998;98:2030–2036[Abstract/Free Full Text]
15. Pires LA, Lehmann MH, Buxton AE, et althe Multicenter Unsustained Tachycardia Trial Investigators. Differences in inducibility and prognosis of in-hospital versus out-of-hospital identified nonsustained ventricular tachycardia in patients with coronary artery disease: Clinical and trial design implications. J Am Coll Cardiol. 2001;38:1156–1162[Abstract/Free Full Text]
16. Newby KH, Thompson T, Stebbins A, et althe GUSTO Investigators. Sustained ventricular arrhythmias in patients receiving thrombolytic therapy: Incidence and outcome. Circulation. 1998;98:2567–2573[Abstract/Free Full Text]
17. Bigger T, Fleiss J, Kleiger R, et al. The relationship among ventricular arrhythmias, left ventricular dysfunction, and mortality in the two years after myocardial infarction. Circulation. 1984;69:250–258[Abstract/Free Full Text]
18. Rahimtoola SH. From coronary artery disease to heart failure: role of the hibernating myocardium. Am J Cardiol 1995;75:16E–22E.
19. Saxon LA, Weiner I, Stevenson WG. Monomorphic versus polymorphic ventricular tachycardia after coronary artery bypass grafting. Am J Cardiol. 1995;75:403–405[CrossRef][Medline]
20. Beauregard LA. Incidence and management of arrhythmias in women. J Gend Specif Med. 2002;5:38–48[Medline]
21. Pham TV, Rosen MR. Sex, hormones, and repolarization. Cardiovasc Res. 2002;53:740–751[Abstract/Free Full Text]
22. Saba S, Zhu W, Aronovitzx MJ, et al. Effects of estrogen on cardiac electrophysiology in female mice. J Cardiovasc Electrophysiol. 2002;13:276–280[CrossRef][Medline]
23. Aronson D, Burger AJ. The effect of sex on ventricular arrhythmic event in patients with congestive heart failure. Pacing Clin Electrophysiol. 2000;25:1206–1211
24. The Amiodarone Trials Meta-Analysis Investigators. Effect of prophylactic amiodarone on mortality after acute myocardial infarction and in congestive heart failure; meta-analysis of individual data from 6500 patients in randomized trials. Lancet. 1997;350:1417–1424[CrossRef][Medline]
25. Exner DV, Reiffel JA, Epstein AE, et al. Beta-blocker use and survival in patients with ventricular fibrillation or symptomatic ventricular tachycardia: The antiarrhythmics versus implantable defibrillators (avid) trial. J Am Coll Cardiol. 1999;34:325–333[Abstract/Free Full Text]
26. Singh BN, Venkatesh N, Nademanee K, et al. The historical development, cellular electrophysiology and pharmacology of amiodarone. Prog Cardiovasc Dis. 1989;31:249–280[CrossRef][Medline]
27. Elwood M. Critical Appraisal of Epidemiological Studies and Clinical Trials. 2nd edition. Oxford: Oxford University Press; 1998.
This article has been cited by other articles:
|
|
|
|
 
F. A. Atik, M. F. M. Garcia, J. M. Baggio Jr, C. N. Faber, R. B. Corso, L. F. Caneo, and A. V. Sarabanda
Recurrent ventricular arrhythmia after coronary artery bypass grafting treated with radiofrequency catheter ablation.
Ann. Thorac. Surg.,
February 1, 2009;
87(2):
631 - 633.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. Chamchad, G. Djaiani, H. J. Jung, L. Nakhamchik, J. Carroll, and J. C. Horrow
Nonlinear Heart Rate Variability Analysis May Predict Atrial Fibrillation After Coronary Artery Bypass Grafting
Anesth. Analg.,
November 1, 2006;
103(5):
1109 - 1112.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
J. P. Singh, W. J. Hall, S. McNitt, H. Wang, J. P. Daubert, W. Zareba, J. N. Ruskin, A. J. Moss, and and the MADIT-II Investigators
Factors Influencing Appropriate Firing of the Implanted Defibrillator for Ventricular Tachycardia/Fibrillation: Findings From the Multicenter Automatic Defibrillator Implantation Trial II (MADIT-II)
J. Am. Coll. Cardiol.,
November 1, 2005;
46(9):
1712 - 1720.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
Top
Abstract
Methods