Article Text
PDF
Abstract
Background Postoperative atrial fibrillation (POAF) affects approximately 30% of patients undergoing elective cardiac surgery. While its pathogenesis is multifactorial, increasing evidence supports a role for oxidative stress in the electrophysiological remodelling associated with AF. Although prophylactic antioxidants appear to be a potentially attractive pharmacotherapy, there is still uncertainty regarding their efficacy. This study aims to provide a quantitative summary of the current evidence surrounding antioxidant vitamins and POAF prevention.
Methods A systematic literature review identified five randomised controlled trials incorporating 567 patients (n=284 antioxidant, n=283 control). These were meta-analysed using random effects modelling. Heterogeneity, subgroup analysis, quality scoring and risk of bias were assessed. Primary endpoints were the incidence of POAF and all-cause arrhythmia. Secondary endpoints were length of stay in the intensive care unit (ITU) and length of hospital stay.
Results Vitamins C and E significantly reduced the incidence of POAF (OR 0.43, 95% CI 0.21 to 0.89) and all-cause arrhythmia (OR 0.54, 95% CI 0.29 to 0.99) compared with controls. A significant reduction in both ITU stay (weighted mean difference (WMD) −0.44, 95% CI −0.70 to −0.17) and hospital stay (WMD −1.11, 95% CI −1.70 to −0.52) was also seen in the antioxidant group, without significant heterogeneity.
Conclusions The prophylactic use of vitamins C and E may significantly reduce the incidence of POAF and all-cause arrhythmia following cardiac surgery. However, the overall quality of current studies is poor and further research should focus on adequately powered randomised controlled trials that standardise AF reporting, antioxidant protocol and the use of concomitant agents. Cost analysis should be considered to establish the potential economic benefit of antioxidant vitamin prophylaxis in POAF.
- Atrial fibrillation
- oxidative stress
- cardiopulmonary bypass
- cardiac surgery
- IHD
- methodology
- metabolic medicine
- paediatric surgury
- atrial fibrillation
- statins
- aortic valve disease
- cardiopulmonary bypass
- surgery-coronary bypass
Statistics from Altmetric.com
- Atrial fibrillation
- oxidative stress
- cardiopulmonary bypass
- cardiac surgery
- IHD
- methodology
- metabolic medicine
- paediatric surgury
- atrial fibrillation
- statins
- aortic valve disease
- cardiopulmonary bypass
- surgery-coronary bypass
Introduction
Atrial fibrillation (AF) is the commonest rhythm disturbance following cardiac surgery, affecting approximately 30–60% of patients.1 2 With appropriate medical therapy, postoperative atrial fibrillation (POAF) is often self-limiting, but it is associated with significant postoperative morbidity and mortality which triples the risk of cardiac-related death and quadruples the risk of disabling embolic events.1 In the short term, POAF results in a significantly longer stay in the intensive care unit (ITU) and in hospital, an increase in resource utilisation and an increase in the financial cost of healthcare.3–5 In the long term, new-onset POAF has a significant negative impact on survival following both valvular and coronary surgery,6 7 with a 10-year mortality up to 48% higher in patients developing POAF compared with those who do not.8 Furthermore, recent evidence suggests that correction of AF may significantly reduce the composite risk of cardiac death and major events.9
Postoperative AF is thought to result from a number of ‘triggering’ and ‘sustaining’ factors acting on a pre-existing pro-arrhythmogenic substrate. There is increasing evidence to implicate oxidative stress and inflammation in this process,3 10 11 and several mechanisms of reactive oxygen species (ROS) generation within the atrial myocyte have been described. While a number of cellular mechanisms protect against oxidant damage in normal myocardium, persistent oxidative stress depletes endogenous antioxidants and ultimately results in oxidative damage. Ischaemia-reperfusion injury following coronary surgery, aortic cross-clamping and cardiopulmonary bypass (CPB) stimulate the production of nitrogen and oxygen free radicals, including superoxide, peroxynitrite, hydroxyl and hydrogen peroxide, and result in depletion of endogenous antioxidants such as glutathione.3 Furthermore, oxidative damage—as demonstrated by the presence of glutathione disulfide—may be seen for up to 24 h following aortic cross-clamp removal.3
Guidelines currently exist in Europe and the USA regarding the pharmacological prophylaxis and management of POAF following cardiac surgery.12 13 At present, β blockers are recommended as first-line pharmacotherapy12 13 with sotalol and amiodarone used as alternative agents.12 However, these pharmacological therapies are subject to a number of limitations including treatment failure, toxicity and pro-arrhythmogenicity. Significant effort has therefore been placed in the development of alternative preventive strategies, and the identification of oxidative stress as a mechanism for AF development has made way for antioxidant vitamins as novel therapeutic agents. Vitamins C and E are chain-breaking antioxidants which scavenge free radicals and terminate the propagation of free radical reactions.14 Vitamin E is pivotal to maintenance of membrane stability, acting to prevent lipid peroxidation, whereas vitamin C independently scavenges water-soluble free radicals and acts synergistically with vitamin E, reducing vitamin E radicals and allowing its regeneration.14
Despite a number of mechanistic studies into the role of antioxidant vitamins in the prevention of POAF, there is continued uncertainty regarding their therapeutic potential and consequently no universally accepted guidelines currently exist. The purpose of this study is therefore to provide a quantitative summary of the available evidence surrounding the use of antioxidant vitamins and POAF prevention and to inform our current practice. The primary aim is to evaluate the role of vitamins C and E in reducing the incidence of POAF, both independently and synergistically. The secondary aim is to address the role of antioxidants in reducing the burden of POAF on healthcare, focusing on surrogate markers of healthcare resource utilisation including length of ITU and hospital stay.
Methods
Literature search
A literature search was performed using PubMed, Ovid, Embase and Cochrane databases using MeSH terms ‘atrial fibrillation’, ‘arrhythmias’, ‘antioxidants’, ‘vitamin C’, ‘ascorbic acid’, ‘vitamin E’, ‘tocopherol’, and ‘vitamin A’. The last date for this search was 1 April 2011. A summary of the search strategy is shown in figure 1.
Search strategy.
Inclusion and exclusion criteria
All articles reporting the use of antioxidant vitamins in the prevention of atrial tachyarrhythmias following all cardiac surgery were included in the review (including those reporting the concomitant use of other anti-arrhythmic pharmacotherapies). Studies were excluded from the review if: (1) inconsistency of data did not allow valid extraction; (2) data were duplicated; and (3) the trial was carried out on animal models.
Based on these criteria, three reviewers (LH, SR and AK) independently selected studies for further examination by reading titles and abstracts of all identified citations. All potentially eligible studies were retrieved in full for further assessment. Any disagreement was resolved by discussion with the senior author (TA).
Data extraction
Three authors (LH, SR and AK) independently extracted the following data from each paper using a standardised spreadsheet: first author; year of publication; study type; number of subjects; study population demographics; type of antioxidant, dose and duration of therapy. Specific outcome data were retrieved where possible for the following: incidence of postoperative AF and all-cause arrhythmia (primary endpoints); ITU stay and hospital stay (secondary endpoints). Meta-analysis was performed in line with recommendations of the Cochrane Collaboration and the Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines.15
Analysis was performed using a random effects model.16 Continuous data were analysed using the weighted mean difference (WMD) as the summary statistic with 95% CIs. The point estimate of the WMD was considered statistically significant at p<0.05 if the 95% CI did not include the value 0. Categorical variables were analysed using the OR. An OR of <1 favoured the treatment group and the point estimate of the OR was considered statistically significant at the p<0.05 level if the 95% CI did not include the value 1.
Heterogeneity
Inter-study heterogeneity was explored using the χ2 statistic, but the I2 value was calculated to quantify the degree of heterogeneity across trials that could not be attributable to chance alone. When I2 was more than 50%, significant statistical heterogeneity was considered to be present.
Subgroup analysis
Sensitivity analysis was performed using the following subgroups: (1) all studies; (2) studies reporting AF or atrial flutter as an isolated outcome (excluding other atrial or ventricular arrhythmias); (3) studies using only vitamin C as antioxidant therapy; (4) study size (n ≥50); (5) studies including coronary artery bypass graft (CABG) surgery only. These subgroups were identified in the protocol before conducting the review and analysis.
Quality scoring and risk of bias assessment
The overall quality of these studies was assessed using the Jadad scoring system.17 This tool places emphasis on the following three areas when defining the quality of a randomised controlled trial (RCT): (1) randomization; (2) double blinding; and (3) description of withdrawals and drop-outs. A score of 1 is given for each of the points described. A further point is obtained where the method of randomisation and/or blinding is given and is appropriate; where it is inappropriate a point is deducted.
Risk of bias assessment was performed in accordance with guidelines outlined in the Cochrane handbook for Systematic reviews or Interventions V.5.1.0.18 Two authors (LH, SR) subjectively reviewed all studies and assigned a value of ‘high’, ‘low’ or ‘unclear’ to the following: (1) selection bias (Was there adequate generation of the randomisation sequence? Was allocation concealment satisfactory?); (2) performance and detection bias (Was there blinding of participants, personnel and outcome assessors?); (3) attrition bias (Were incomplete outcome data sufficiently assessed and dealt with?); (4) reporting bias (Was there evidence of selective outcome reporting?); and (5) Were any other sources of bias identified?
The analysis was conducted using Review Manager V.5.0 for Windows (The Cochrane Collaboration, Software Update, Oxford, UK).
Results
Eligible studies, quality and bias assessment
Six publications19–24 were identified as fulfilling the inclusion criteria of this review, producing a pooled data set of 653 patients. Five were RCTs20–24 and were included in the meta-analysis (total 567 patients). The other was prospective but non-randomised.19 The overall quality of these trials was poor as determined by the Jadad score. One study scored highly with a Jadad score of 424 but, of the remaining studies, two21 22 were given a score of 2 and three19 20 23 were given a score of 1 out of a possible 5.
Risk of bias analysis revealed several areas of concern (figure 2), particularly regarding reporting bias in two studies where ventricular and atrial arrhythmias were reported as a single outcome.21 23 In addition to this, only two of the included studies21 24 adequately reported the randomisation protocol and none described a method used to conceal the allocation sequence in sufficient detail to exclude selection bias.
Risk of bias analysis. (A) Risk of bias summary: the authors' judgements about each risk of bias item for the included studies. (B) Risk of bias graph: the authors' judgements about each risk of bias item presented as percentages across all included studies.
Three studies20 22 23 included patients undergoing CABG only, two reported isolated administration of vitamin C as the sole intervention,20 22 four administered concomitant β blocker therapy20–22 24 and three reported only AF or atrial flutter as the arrhythmia outcome.20 22 24 The characteristics of the included studies are shown in table 1.
Synopsis of studies
Intraoperative parameters
Analysis of intraoperative parameters revealed no significant difference between the treatment and placebo groups in number of distal anastomoses performed, CPB time or cross-clamp time (p=0.08, p=0.19 and p=0.86, respectively) without significant heterogeneity (table 2).
Overall results and subgroup analysis
Primary endpoints
The overall results for both primary and secondary endpoints are shown in table 2.
Incidence of POAF and all-cause arrhythmia
The incidence of POAF and/or atrial flutter was reported by three studies (183 antioxidant, 182 control).20 22 24 A significantly lower incidence of AF/atrial flutter and all-cause arrhythmia was seen in the antioxidant group compared with controls, without significant heterogeneity (AF/atrial flutter: OR 0.43, 95% CI 0.21 to 0.89, p=0.02; all-cause arrhythmia: five studies (n=567), OR 0.54, 95% CI 0.29 to 0.99, p=0.05; figure 3).
Forest plots showing the treatment effect of antioxidants on the incidence of postoperative atrial fibrillation (AF). (A) Atrial fibrillation only. (B) All-cause arrhythmia.
Subgroup analysis
Three studies20 22 23 reported the incidence of arrhythmia in patients undergoing CABG only (216 antioxidant, 216 control). A significantly lower incidence of all arrhythmias was seen in the antioxidant group compared with controls (OR 0.39, 95% CI 0.20 to 0.77, p=0.006), without significant heterogeneity.
No significant difference was seen between the antioxidant and control groups when studies administering vitamin C as the sole antioxidant or larger studies (n ≥50) were analysed. Notably, both subgroups are limited to the same two trials (135 control, 135 antioxidant)20 22 and moderate heterogeneity is present between them (χ2 3.03, p=0.08, I2 67%). Furthermore, both of these studies also concomitantly administered β blockers to all patients.
Secondary endpoints
Length of stay
Three studies20–22 reported length of both ITU and hospital stay. One further study reported hospital stay only.24 A significant reduction in ITU stay (WMD −0.44, 95% CI −0.70 to −0.17, p=0.001) and hospital stay (WMD −1.11, 95% CI −1.70 to −0.52, p=0.0002) was seen in the antioxidant group compared with controls, without significant heterogeneity.
Discussion
Our results quantitatively demonstrate a reduction in the incidence of POAF and all-cause arrhythmia following antioxidant vitamin therapy which remained significant when only RCTs were analysed and was independent of surgical procedure type. Furthermore, no significant difference in CPB time or cross-clamp time was observed between the antioxidant and control groups, suggesting that this did not contribute to our observed reduction in POAF and arrhythmia. These findings are further corroborated by the work of Carnes et al who, in their prospective study of 50 patients, reported a 16.3% incidence of POAF in ascorbate-treated patients compared with 34.9% in controls.19
Evidence also exists to suggest a synergistic effect between β blockers and antioxidant vitamins in reducing the incidence of POAF.20 While the exact mechanism for this apparent synergism is yet to be elucidated, it is thought that heightened sympathetic tone worsens oxidative stress and may potentiate Ca2+ overload. We note that one study included in this review did not comment on matching for concomitant β blocker usage,23 however exclusion of this study from the analysis did not affect the observed reduction in either POAF or all-cause arrhythmia seen with antioxidant therapy.
Role of antioxidants in POAF
The potential role of oxidative stress in the initiation and maintenance of AF is becoming increasingly recognised. Peroxynitrite and hydroxyl radicals been identified in the atrial appendages of patients with chronic AF,10 and AF has also been associated with a shift in gene expression, promoting genes coding for ROS.25 In their canine model of AF, Carnes and colleagues go on to link this oxidative stress to electrical atrial remodelling. Using rapid in vivo pacing of canine atria, they observed an increase in markers of peroxynitrite and a reduction in the effective refractory period.19
As mounting evidence for the role of oxidative stress in the pathogenesis of AF has become apparent, antioxidant vitamins have been explored as both preventive and therapeutic agents. Vitamins C and E are chain-breaking antioxidants acting on water- and lipid-soluble zones, respectively, to scavenge free radicals and terminate the propagation of free radical reactions. Their differing solubility enables synergistic ROS scavenging and also regeneration of vitamin E by vitamin C at the lipid-water interface.14 26 Carnes and colleagues demonstrated a reduction in atrial tissue levels of vitamin C, reflecting increased consumption following rapid atrial pacing. Moreover, exogenous administration of vitamin C both attenuated this pacing-induced reduction in effective refractory period and blunted the production of peroxynitrite.19
Impact of antioxidant therapy on length of stay and the economic burden of AF
POAF is a well-recognised cause of prolonged hospital stay and increased financial burden after cardiac surgery.4 27 28 Our results demonstrate a reduction in both ITU and overall hospital stay in the antioxidant group compared with controls. While we recognise that this cannot be attributed to a reduction in POAF alone, it is likely that a reduction in POAF following antioxidant administration may significantly contribute to this finding.
Cost analysis of POAF has estimated its overall financial burden to range from $2574 to $6356 per patient.27 29 This reflects increased resource utilisation in several areas including ward and ITU consumables, nursing care, laboratory fees, imaging, pharmacotherapy and anaesthetic support.27 Other pharmacological agents such as amiodarone have not been shown to be cost-effective in the prophylaxis of POAF following CABG, either in the short or long term.30 In addition, the use of amiodarone and other anti-arrhythmic drugs is often limited by potential side effects, toxicity, drug interactions and intolerance. In contrast, vitamins C and E appear to be safe and are associated with few systemic side effects. We therefore hypothesise that prophylaxis with antioxidant vitamins may prove to be safe and cost-effective in the prevention of POAF.31
Study limitations
The primary aim of this review is to provide an evidence base to guide preoperative planning while further work is carried out. However, our results are subject to a number of limitations. First, our analysis is based on only five RCTs, of which only two are double-blind. In addition, most of these studies are small (n≤50 in three studies), of low quality (four studies with Jadad score ≤2) and are underpowered. Based on the current data, power calculation suggests 355 patients are necessary in each arm to enable detection of a 10% reduction in POAF (at an estimated occurrence rate of 25%), with 90% power.
Second, a mixture of vitamins C and E were used in differing doses and often in conjunction with other agents such as β blockers and allopurinol. We have attempted to address this by performing subgroup analysis to isolate the independent effect of vitamin C and vitamin E, with and without concomitant β blockade, and have produced comparable results, although we acknowledge that these results are limited by the small size of the subgroups.
Third, these studies lack homogeneity in both the method of postoperative monitoring and in their definition of POAF. While no clear guidelines exist on what constitutes an episode of POAF following cardiac surgery, the Heart Rhythm Society expert consensus guidelines for surgical ablation suggest that AF should be documented if it is recorded for >30 s on Holter monitoring.32 Not only do these studies vary the method of AF detection from daily ECGs to continuous monitoring, but the recording trigger varies from episodes lasting >1 min to those lasting >10 min. This leads to potential underestimation and/or overestimation of the true incidence of POAF.
Finally, we have identified several areas of potential bias within the studies included in this review. It is important to note that only published data were included in this meta-analysis, and it is possible that the true effect of antioxidants is over-represented owing to under-publication of smaller trials or those reporting negative findings. Also of particular concern is a high risk of reporting bias where atrial and ventricular arrhythmias are expressed as a combined outcome.21 23 While electrolyte imbalance, increased adrenergic tone and the existence of a pre-existing pro-arrhythmogenic substrate may be common predisposing factors to both atrial and ventricular arrhythmias, their aetiology is often very different. Acute myocardial ischaemia resulting from graft failure or inadequate myocardial protection may precipitate ventricular arrhythmias both intraoperatively and postoperatively. Furthermore, ventricular arrhythmias may be triggered by scar tissue and hibernating myocardium in patients with previous ischaemic injury,33 34 factors unlikely to be influenced by the administration of antioxidant therapy.
In an attempt to address the issue of publication bias and identify ongoing research, we performed a search of the http://ClinicaTrials.gov trial registry. Three trials exploring the role of ascorbic acid and atrial fibrillation (NCT00519337, NCT00953212 and NCT01107730) were identified, registered in 2008, 2009 and 2010, respectively. These were distinct from the published trials included in this meta-analysis. Two of these studies are ongoing or in the recruitment phase (NCT00953212 and NCT01107730) and the status of the other is unknown. While all incorporate a randomised double-blind study design, two are of small sample size (estimated number of patients 50 and 87, respectively) and the other (estimated number of patients 304) includes four arms with multiple interventions. These studies will thus again be underpowered and heterogeneity will remain a problem when carrying out pooled analysis.
Future research should therefore focus on an adequately powered multicentre double-blind placebo-controlled randomised trial and include an appropriately designed cost-effectiveness analysis. Furthermore, novel studies should aim to standardise AF reporting criteria and antioxidant protocol, independently isolating each antioxidant before investigating the potential benefits of concomitant agents and antioxidant vitamin synergism.
Conclusion
This meta-analysis suggests that prophylactic use of antioxidant vitamins C and E significantly reduces the incidence of POAF and all-cause arrhythmia following cardiac surgery. Furthermore, as a result of a concomitant reduction in both ITU and hospital stay, we hypothesise that antioxidant therapy may potentially reduce hospital resource utilisation and financial cost. Further research should focus on adequately powered RCTs, standardising AF reporting criteria, antioxidant protocol and the use of concomitant agents, and should also consider cost analysis to establish the potential economic benefit of antioxidant vitamin therapy as routine prophylaxis in POAF.
References
Footnotes
Competing interests None.
Ethics approval Only published studies with their own independent ethical approval were included in this meta-analysis.
Provenance and peer review Not commissioned; externally peer reviewed.