Treatment strategies for atrial fibrillation after cardiothoracic surgery

National Guideline Centre (UK)

Treatment strategies for atrial fibrillation after cardiothoracic surgery

Atrial fibrillation: diagnosis and management

Evidence review L

NICE Guideline, No. 196

Authors

National Guideline Centre (UK).

London: National Institute for Health and Care Excellence (NICE); 2021 Apr.

1. Treatment strategies for atrial fibrillation after cardiothoracic surgery

1.1. Review question: What is the most clinical and cost effective treatment strategy (rate or rhythm control,or no treatment) for peoplewith atrial fibrillation after cardiothoracic surgery?

1.2. Introduction

Atrial fibrillation remains one of the most common adverse events to occur following cardiac surgery. Despite the improvement in the rate of other perioperative morbidities and mortality, however, the reported incidence of post-operative AF following cardiac surgery remains high (up to 30–50%) and has not changed significantly over recent years. Its incidence increases in those with an increased age, undergoing surgery of increased complexity or with a past history of AF, and is associated with a significant increase in perioperative morbidity, hospital length of stay, utilisation of health care resources and mortality.

The exact mechanisms of initiation and maintenance of post-cardiac surgery AF, however, are not fully understood and associated with this, a number of different treatment modalities and strategies (rate or rhythm control) have been proposed. A rate control strategy includes using medications (such as beta-blockers or calcium channel blockers) that reduce conduction across the atrioventricular node to slow the heart rate, whereas rhythm control strategies include using pharmacological agents (such as amiodarone) or electrical cardioversion in an attempt to restore sinus rhythm. Other considerations for these patients include identification and treatment of any triggers of atrial fibrillation, such as restoration of serum potassium and magnesium levels; anticoagulation, whilst balancing bleeding risks with thromboembolic risks; and the haemodynamic status of the patient, where early electrical cardioversion may be required in patients with hypotension or marked tachycardia.

Due to the absence of robust clinical studies, the implementation of these different management strategies varies considerably. This chapter intends to examine the clinical evidence surrounding the different therapeutic options used in the treatment of atrial fibrillation following cardiac surgery and develop some recommendations regarding how best to manage these patients.

1.3. PICO table

For full details see the review protocol in Appendix A:.

Table 1

PICO characteristics of review question.

1.4. Methods and process

This evidence review was developed using the methods and process described in Developing NICE guidelines: the manual.⁸³Methods specific to this review question are described in the review protocol in Appendix A:.

1.5. Clinical evidence

1.5.1. Included studies

A search was conducted to identify randomised controlled trials or systematic reviews of randomised controlled trials comparing different strategies for treating atrial fibrillation after cardiothoracic surgery, including rate control (beta-blockers, calcium channel blockers, digoxin and amiodarone), rhythm control (Na+ blockers, K+ blockers and DC cardioversion) and no treatment strategies. The population could include those developing atrial fibrillation after surgery and also those with pre-existing atrial fibrillation prior to the surgery – population strata were used to stratify for the presence or absence of pre-existing atrial fibrillation prior to surgery from the outset of the review. Fifteen studies (from sixteen papers) were included in the review;¹²^, ¹³^, ¹⁵^, ³¹^, ³⁵^, ³⁷^, ⁵¹^, ⁶⁴^, ⁶⁷^, ⁶⁸^, ⁸⁴^, ⁸⁹^, ¹⁰⁰^, ¹⁰¹^, ¹¹⁰^, ¹¹¹ these are summarised in Table 2 below. Evidence from these studies is summarised in the clinical evidence summary below (Tables 3-17).

See also the study selection flow chart in Appendix C:, study evidence tables in Appendix D:, forest plots in Appendix E:andGRADE tables in Appendix F:.

The majority of studies (eleven studies from twelve papers) included in this review were within the no pre-existing atrial fibrillation stratum, where in most cases the presence of atrial fibrillation prior to the cardiothoracic surgery was an exclusion criterion. Of the remaining four papers, three were within the pre-existing atrial fibrillation stratum and one was assigned to the mixed/unclear stratum as there were limited details to assign it to one of the two other strata.

The included studies covered the following comparisons between the interventions listed in the protocol for this review:

Mixed/unclear stratum:

One study compared calcium channel blockers (intravenous verapamil) with intravenous placebo;³⁷

No pre-existing AF stratum:

One study compared DC cardioversion with K+ blockers (intravenous amiodarone);³¹
One study compared K+ blockers (intravenous amiodarone) with intravenous digoxin;¹⁵
Two studies compared K+ blockers (intravenous followed by oral amiodarone) with K+ blockers + ranolazine (intravenous amiodarone and oral ranolazine, followed by oral amiodarone and oral ranolazine);¹⁰⁰^, ¹⁰¹
One study compared a mixed rate control strategy (specific drugs not stated, oral administration likely but not clear) with K+ blockers (amiodarone recommended, oral administration) +/-a rate control agent (specific drugs not stated, oral administration likely but not clear);³⁵
One study (covered by two papers) compared a mixed rhythm control strategy (such as sotalol, propafenone or procainamide, route of administration dependent on drug used) +/- electrical cardioversion with a mixed rate control strategy (such as diltiazem, verapamil, metoprolol, atenolol, propranolol, esmolol or digoxin - route of administration dependent on drug used and patient);⁶⁷^, ⁶⁸
One study compared Na+ blockers (intravenous flecainide) with intravenous digoxin¹¹¹
One study compared Na+ blockers (oral propafenone) with K+ blockers (intravenous amiodarone);⁸⁴
One study compared calcium channel blockers(intravenous verapamil)with intravenous placebo;⁵¹
One study compared K+ blockers (intravenous vernakalant) with intravenous placebo;⁶⁴
One study compared K+ blockers (amiodarone) with routine medical treatment alone;¹²

Pre-existing AF stratum:

One study compared DC cardioversion (oral amiodarone for four days prior to cardioversion) with K+ blockers + captopril + simvastatin (oral amiodarone with captopril and simvastatin);¹³
One study compared a mixed rate control strategy (digoxin and/or diltiazem, route unclear but likely to be oral based on study length) with K+ blockers + captopril + simvastatin(oral amiodarone with captopril and simvastatin);⁸⁹
One study compared K+ blockers(intravenous followed by oral amiodarone) + DC cardioversion with placebo (unclear if given intravenously followed by orally as with amiodarone) + DC cardioversion;¹¹⁰

It is also noted that studies which included intravenous use of certain drugs (including diltiazem, sotalol, disopyramide and propafenone) that are only available in the UK in oral form and not used intravenously were not included in the review. One study(from two papers)that compared a mixed rhythm control strategy +/- electrical cardioversion with a mixed rate control strategy was downgraded for intervention indirectness, as one of the options within the mixed rate control strategy was the use of intravenous diltiazem but the proportion of patients that received this as their rate control strategy was unclear⁶⁷^, ⁶⁸.

1.5.2. Excluded studies

See the excluded studies list in Appendix I:.

1.5.3. Summary of clinical studies included in the evidence review

Table 2

Summary of studies included in the evidence review.

See Appendix D:for full evidence tables.

1.5.4. Quality assessment of clinical studies included in the evidence review

Table 3

Clinical evidence summary: Evidence not suitable for GRADE analysis.

Table 4

Clinical evidence summary: Mixed/unclear stratum– calcium channel blockers vs. placebo.

Table 5

Clinical evidence summary: No pre-existing AFstratum– DC cardioversion vs. K+ blockers.

Table 6

Clinical evidence summary: No pre-existing AFstratum– K+ blockers vs. digoxin.

Table 7

Clinical evidence summary: No pre-existing AF stratum– K+ blockers vs. K+ blockers + ranolazine.

Table 8

Clinical evidence summary: No pre-existing AF stratum– mixed rate control vs. K+ blockers with/without rate control agent.

Table 9

Clinical evidence summary: No pre-existing AF stratum– mixed rhythm control +/- electrical cardioversion vs. mixed rate control.

Table 10

Clinical evidence summary: No pre-existing AF stratum– Na+ blockers vs. digoxin.

Table 11

Clinical evidence summary: No pre-existing AF stratum– Na+ blockers vs. K+ blockers.

Table 12

Clinical evidence summary: No pre-existing AFstratum– Calcium channel blockers vs. placebo.

Table 13

Clinical evidence summary: No pre-existing AFstratum– K+ blockers (vernakalant) vs. placebo.

Table 14

Clinical evidence summary: No pre-existing AF stratum– K+ blockers (amiodarone) vs. routine medical treatment alone.

Table 15

Clinical evidence summary: Pre-existing AF stratum– DC cardioversion vs. K+ blockers + captopril + simvastatin.

Table 16

Clinical evidence summary: Pre-existing AF stratum– Mixed rate control vs. K+ blockers + captopril + simvastatin.

Table 17

Clinical evidence summary: Pre-existing AF stratum– K+ blockers + DC cardioversion vs. placebo + DC cardioversion.

See Appendix F: for full GRADE tables.

Click here to enter text.

1.6. Economic evidence

1.6.1. Included studies

No health economic studies were included.

1.6.2. Excluded studies

No relevant health economic studies were excluded due to assessment of limited applicability or methodological limitations.

See also the health economic study selection flow chart in Appendix G:.

1.6.3. Unit costs

Relevant drug unit costs are provided in Table 18 to aid consideration of costeffectiveness.

Note, the Na+ channel blocker procainamide is only available from ‘special-order’ manufacturers or specialist importing companies and so has not been costed below.

Table 18

Drug unit costs.

Direct current cardioversion (X501) and external cardioversion electrical cardioversion (X502) are not coded separately as a HRG, and therefore the day case unit cost for Arrhythmia or Conduction Disorders (EB07) is the closest proxy, which has a weighted cost of £670taking comorbidities and or complications into account.²²

The weighted average cost for excess bed days for patients who have had elective and non-elective CABG are provided in Table 19 and Table 20.

Table 19

Elective inpatient excess bed days cost.

Table 20

Non-elective inpatient excess bed days cost.

1.7. The committee’s discussion of the evidence

1.7.1. Interpreting the evidence

1.7.1.1. The outcomes that matter most

In this review, the following outcomes were considered to be critical for decision-making: health-related quality of life, mortality, stroke or thromboembolic complications, need for rescue DC cardioversion, rehospitalisation (all-cause), rehospitalisation for AF, achievement of sinus rhythm and adverse events.

Additional outcomes that were considered to be important for decision-making were freedom from anticoagulation, freedom from antiarrhythmic drug use, hospital length of stay and intensive care unit length of stay.

In this review, no clinical evidence was identified for the following critical outcomes: freedom from antiarrhythmic drug use and intensive care unit length of stay.

1.7.1.2. The quality of the evidence

The quality of the evidence for all outcomes included in this review ranged from very low quality to moderate, with the majority of outcomes for all comparisons being rated very low quality based on GRADE quality assessment. For those rated very low quality, the main factors contributing to the quality rating were a very high or high risk of bias and imprecision in the effect estimates due to very small study sample sizes. There were only four studies where the population size was >100 participants, and the number of participants within the other nine studies ranged from 14 to 99, which made imprecision an issue with the majority of outcomes from these smaller studies. This made it difficult for the committee to interpret these outcomes and decide whether any clinically important differences were present due to the uncertainty surrounding the effect estimates. For the majority of outcomes only one study was available and meta-analysis was therefore not performed.

In addition, there were some outcomes presented that could not be analysed and assessed by GRADE due to insufficient detail reported within the studies. This comprised two outcomes, one where only the median and interquartile range was given for the length of hospital stay in both groups and the other where data was not given for separately for each randomised group. These outcomes were reported separate to other outcomes and are presented in Table 3 of the evidence review.

1.7.1.3. Benefits and harms

The evidence included in this review was obtained from thirteen RCTs (covered by fourteen papers) and was stratified from the outset based on whether or not the AF was pre-existing before cardiothoracic surgery was performed within the individual studies.

For the no pre-existing AF stratum, where AF was new-onset following cardiothoracic surgery, nine RCTs were included, which covered the following comparisons: DC cardioversion vs. K+ blockers, K+ blockers vs. digoxin, K+ blockers vs. K+ blockers + ranolazine, mixed rate control vs. K+ blockers with/without rate control, mixed rhythm control with/without electrical cardioversion vs. mixed rate control, Na+ blockers vs. digoxin, Na+ blockers vs. K+ blockers and calcium channel blockers vs. placebo.

For the pre-existing AF stratum, where AF was present prior to the cardiothoracic surgery, three RCTs were included, which covered the following comparisons: DC cardioversion vs. K+ blockers + captopril + simvastatin, mixed rate control vs. K+ blockers + captopril + simvastatin and K+ blockers + DC cardioversion vs. placebo + DC cardioversion.

There was an additional study that could not be classified into either of the above strata as there was no information about preoperative AF – this was included separately under a mixed/unclear stratum and covered the comparison between calcium channel blockers and placebo.

No pre-existing AF

For the majority of the evidence within this stratum, the committee agreed that there was insufficient evidence to favour particular interventions and that many of the studies were old with very small participant numbers and covered drugs that are not commonly used in practice anymore. For the comparisons between individual drug classes (such as Na+ blockers vs. K+ blockers) there was only one, small study for each and there was substantial uncertainty in the effect estimate for most outcomes due to imprecision, meaning the committee felt that there was insufficient evidence to favour a particular drug class.

However, the committee noted the inclusion of a larger RCT that compared a mixed rate control strategy vs. K+ blockers (amiodarone) with/without rate control that reported numerous outcomes listed in the protocol, all of which appeared to suggest no clinical difference (or there was uncertainty around the effect estimate and the true effect size) between the two groups, particularly concerning the presence of sinus rhythm at discharge, freedom from warfarin at discharge and all-cause hospital readmission at 60 days. There was also no difference in hospital length of stay based on median values reported in the study.

There were some outcomes where the point estimate of the relative effect suggested a benefit of K+ blockers with/without rate control (mortality and stroke or thromboembolic complications); however the committee did not consider these to be clinically important differences based on the absolute effect estimates, the presence of imprecision and the low number of events. The adverse event outcome (serious and non-serious events) reported in this study appeared to suggest a slight benefit of mixed rate control, which the committee agreed may be due to the side effects associated with amiodarone use, however this was also considered not to be a clinically important difference based on the size of the effect. Need for rescue DC cardioversion also suggested a benefit of mixed rate control over amiodarone treatment, however, DC cardioversion was a recommended procedure in the amiodarone group if patients did not respond (and this was not mentioned within the mixed rate control group procedure), which may partially explain the increased number in the amiodarone group.

Although this RCT was not without its limitations, such as no details provided about the types of rate control drugs included in the mixed rate control group, the committee noted that this was the best available evidence within the review to inform changes to the existing recommendation covering the post-cardiothoracic surgery population.

Overall, the committee considered that for those with no pre-existing AF prior to cardiothoracic surgery, there was insufficient evidence to keep the existing strong recommendation to offer a rhythm control strategy as the initial management option for postoperative atrial fibrillation following cardiothoracic surgery, and therefore agreed that this should be changed to a consider recommendation, which would give less emphasis on rhythm control strategies and allow rate control strategies to be considered if the clinician felt this was more appropriate for the individual patient. The committee agreed that this may lead to a change in practice as in their experience the use of amiodarone to treat new-onset AF following cardiothoracic surgery is routine. The committee noted that this routine use of amiodarone may be unnecessary based on the results of the review and the possibility that postoperative AF may resolve naturally in many patients with watchful waiting, meaning that a rate control strategy rather than a rhythm control strategy may often be sufficient to resolve the atrial fibrillation. This contributed to the committee’s decision to change the recommendation to a consider recommendation. The committee also noted that if a rate control strategy was initially selected instead of a rhythm control strategy, rhythm control would remain an option if this initial management failed. A benefit of a watchful waiting strategy could include avoiding side effects associated with rate and rhythm control drugs (particularly amiodarone), though side effects are considered to be less of an issue with the short term use of these drugs. However, limited evidence was included in this review comparing rate or rhythm control drugs with a watchful waiting strategy, and therefore no recommendation was made concerning this strategy.

The committee agreed that if a rhythm control strategy was chosen as the initial management, the need for this should be reviewed, alongside the need for any associated anticoagulation, at approximately 6 weeks and not continued automatically for long periods of time. 6 weeks is in line with current practice and is an appropriate time point to assess the person’s recovery including for example prosthetic valve function and to check if sinus rhythm has been restored. The committee agreed that the adverse events associated with the use of amiodarone are usually following medium to long-term use and were less concerned about these adverse events for the treatment of new-onset AF following cardiothoracic surgery, providing amiodarone use is not continued for long periods unnecessarily.

Pre-existing AF

The committee agreed that there was insufficient evidence within this stratum to make any specific recommendations for those with pre-existing AF before cardiothoracic surgery that remained following surgery. The evidence was obtained from three studies, which covered three separate comparisons.

There was some evidence from one study that DC cardioversion improved clinical outcomes compared with a group where K+ blockers were used, and evidence from another study that suggested K+ blockers with DC cardioversion was better than DC cardioversion alone in terms of achieving sinus rhythm. However, these studies were substantially smaller than the largest RCT included for the no pre-existing AF stratum and the committee felt unable to make recommendations based on his.

In addition to the lack of evidence, the committee noted that all three studies in this stratum covered the patients that were undergoing mitral valve surgery and that in this situation most patients with pre-existing AF prior to mitral valve surgery would undergo simultaneous left atrial surgery at the time of valve intervention, with the aim of resolving the AF and reducing the need for future intervention or treatment.

1.7.2. Cost effectiveness and resource use

No relevant health economic analyses were identified for this review. Relevant unit costs were presented for rate and rhythm strategies. The unit cost of rate and rhythm drugs was generally low and comparable. Although there were some more costly drugs within each class, these were considered by the committee to be less frequently used than the other lower cost drugs. This was because the low cost drugs generally work well in terms of acute conversion and there is no evidence of any gains of using the more expensive drugs currently. Furthermore, the aim of this review question was to compare rate versus rhythm strategies rather than making inter class comparisons, therefore this was not considered an issue. Finally, the committee noted that rate or rhythm drugs would be used for a short period of time in this context, usually over a period of days or weeks. The unit cost of direct current cardioversion was also presented. Direct current cardioversion (X501) and external cardioversion electrical cardioversion (X502) are not coded separately as an HRG, and therefore the day case unit cost for Arrhythmia or Conduction Disorders (EB07) is the closest proxy, which has a weighted cost of £670 taking comorbidities and or complications into account. The committee noted that the cost would likely be lower when done in the intensive care unit or in an outpatient setting. Finally, the unit cost of excess bed days for patients undergoing CABG was presented to illustrate the potential cost of strategies that increase length of stay.

The committee considered these unit costs alongside the clinical evidence summarised above. For people with no pre-existing atrial fibrillation, they agreed that the limited clinical evidence available did not support the previous strong recommendation to offer rhythm control. Instead they made a weaker ‘consider’ recommendation reflecting the limited clinical evidence and lack of economic evidence. This was based primarily on a single large RCT which showed no difference between rate and rhythm strategies for a number of the protocol outcomes. This study included outcomes associated with resource use such as rehospitalisation and rescue direct current cardioversion. For all cause rehospitalisation, there was no difference between rate and rhythm strategies. For rescue cardioversion and AF rehospitalisation outcomes, they favoured rate control, however there was uncertainty surrounding the point estimate. Overall it was considered that this amendment to the recommendation may reduce the use of rhythm control with drugs such as amiodarone, and increase the use of rate control drugs. As the cost of the drugs is similar and there is no reported significant difference in downstream resource use from the clinical evidence, it was thought that this recommendation amendment is unlikely to have a significant resource impact.

For those with pre-existing AF, no specific recommendation was made for this population due to a lack of robust and relevant clinical evidence.

1.7.3. Other factors the committee took into account

The committee also considered the role of patient preference in the decision to use a rhythm or rate control strategy following the development of new-onset AF post-cardiothoracic surgery. The committee agreed that where possible the clinician would discuss with the patient their preferences, but also noted that in many cases the patient would be acutely unwell and unable to communicate their preferences to the clinician, meaning the clinician would have to make the decision without patient input in these cases.

References

1.: Anonymous. Managing atrial fibrillation after coronary artery bypass graft surgery involves prophylaxis, cardioversion and/or ventricular rate control. Drugs and Therapy Perspectives. 2006; 22(9):12–15

2.: Balser JR, Martinez EA, Winters BD, Perdue PW, Clarke AW, Huang W et al. Beta-adrenergic blockade accelerates conversion of postoperative supraventricular tachyarrhythmias. Anesthesiology. 1998; 89(5):1052–1059 [PubMed: 9821992]

3.: Banach M, Okonski P, Zaslonka J. Atrial fibrillation following cardiosurgical operations - current guidelines of pharmacotherapy and invasive treatment. Polski Przeglad Chirurgiczny. 2005; 77(4):398–412

4.: Bechtel JF, Christiansen JF, Sievers HH, Bartels C. Low-energy cardioversion versus medical treatment for the termination of atrial fibrillation after CABG. Annals of Thoracic Surgery. 2003; 75(4):1185–1188 [PubMed: 12683560]

5.: Bernard EO, Schmid ER, Schmidlin D, Scharf C, Candinas R, Germann R. Ibutilide versus amiodarone in atrial fibrillation: a double-blinded, randomized study. Critical Care Medicine. 2003; 31(4):1031–1034 [PubMed: 12682468]

6.: Blessberger H, Lewis S, Pritchard M, Fawcett L, Domanovits H, Schlager O et al. Perioperative beta-blockers for preventing surgery-related mortality and morbidity in adults undergoing cardiac surgery. Cochrane Database of Systematic Reviews 2019, Issue 9. Art. No.: CD013435. DOI: 10.1002/14651858.CD013435. [PMC free article: PMC6755267] [PubMed: 31544227] [CrossRef]

7.: BMJ Group and the Royal Pharmaceutical Society of Great Britain. British National Formulary. Available from: https://www.evidence.nhs.uk/formulary/bnf/current Last accessed: 21/01/2020.

8.: Bockeria OL, Kanametov TN, Shvartz VA, Sokolskaya MA, Zhuginisov DS, Sanakoev MK et al. Epicardial application of hydrogel with amiodarone for prevention of postoperative atrial fibrillation in patients after coronary artery bypass grafting. Journal of Cardiovascular Translational Research. 2020; 13: 191–198 [PubMed: 31367901]

9.: Burke SW, Solomon AJ. Atrial fibrillation in patients after cardiovascular surgery: incidence, risk factors, preventive and therapeutic strategies. American Journal of Cardiovascular Drugs. 2003; 3(2):95–100 [PubMed: 14727936]

10.: Campbell TJ, Gavaghan TP, Morgan JJ. Intravenous sotalol for the treatment of atrial fibrillation and flutter after cardiopulmonary bypass. Comparison with disopyramide and digoxin in a randomised trial. British Heart Journal. 1985; 54(1):86–90 [PMC free article: PMC481854] [PubMed: 3893488]

11.: Campbell TJ, Morgan JJ. Treatment of atrial arrhythmias after cardiac surgery with intravenous disopyramide. Australian and New Zealand Journal of Medicine. 1980; 10(6):644–649 [PubMed: 6938185]

12.: Chen K, Qin L, Lu X, Xia T, Gu Q. Amiodarone in the treatment of atrial fibrillation of patients with rheumatic heart disease after valve replacement. Pakistan Journal of Medical Sciences. 2019; 35(4):918–922 [PMC free article: PMC6659075] [PubMed: 31372117]

13.: Chen WS, Gao BR, Chen WQ, Li ZZ, Xu ZY, Zhang YH et al. Comparison of pharmacological and electrical cardioversion in permanent atrial fibrillation after prosthetic cardiac valve replacement: a prospective randomized trial. Journal of International Medical Research. 2013; 41(4):1067–1073 [PubMed: 23794519]

14.: Cheng JW. Vernakalant in the management of atrial fibrillation. Annals of Pharmacotherapy. 2008; 42(4):533–542 [PubMed: 18334607]

15.: Cochrane AD, Siddins M, Rosenfeldt FL, Salamonsen R, McConaghy L, Marasco S et al. A comparison of amiodarone and digoxin for treatment of supraventricular arrhythmias after cardiac surgery. European Journal of Cardio-Thoracic Surgery. 1994; 8(4):194–198 [PubMed: 8031562]

16.: Connolly SJ, Mulji AS, Hoffert DL, Davis C, Shragge BW. Randomized placebo-controlled trial of propafenone for treatment of atrial tachyarrhythmias after cardiac surgery. Journal of the American College of Cardiology. 1987; 10(5):1145–1148 [PubMed: 3499456]

17.: Conte SM, Florisson DS, De Bono JA, Davies RA, Newcomb AE. Management of atrial fibrillation after cardiac surgery. Internal Medicine Journal. 2019; 49(5):656–658 [PubMed: 31083803]

18.: Crystal E, Kahn S, Roberts R, Thorpe K, Gent M, Cairns JA et al. Long-term amiodarone therapy and the risk of complications after cardiac surgery: results from the Canadian Amiodarone Myocardial Infarction Arrhythmia Trial (CAMIAT). Journal of Thoracic and Cardiovascular Surgery. 2003; 125(3):633–637 [PubMed: 12658206]

19.: Dagres N, Iliodromitis EK, Lekakis JP, Simitsis P, Karatzas D, Rallidis LS et al. Ranolazine for the prevention or treatment of atrial fibrillation: a systematic review. Journal of Cardiovascular Medicine. 2014; 15(3):254–259 [PubMed: 24662415]

20.: Daoud EG. Management of atrial fibrillation in the post-cardiac surgery setting. Cardiology Clinics. 2004; 22(1):159–166 [PubMed: 14994855]

21.: De Vecchis R, Ariano C, Giasi A, Cioppa C. Antiarrhythmic effects of ranolazine used both alone for prevention of atrial fibrillation and as an add-on to intravenous amiodarone for its pharmacological cardioversion: a meta-analysis. Minerva Cardioangiologica. 2018; 66(3):349–359 [PubMed: 28497941]

22.: Department of Health. NHS reference costs 2017–18. 2018. Available from: https://improvement.nhs.uk/resources/reference-costs/#rc1718 Last accessed: 21/01/20.

23.: deSouza IS, Tadrous M, Sexton T, Benabbas R, Carmelli G, Sinert R. Pharmacologic cardioversion of recent-onset atrial fibrillation and flutter in the emergency department: a systematic review and network meta-analysis. Annals of Emergency Medicine. 2020; 76(1):14–30 [PubMed: 32173135]

24.: deSouza IS, Tadrous M, Sexton T, Benabbas R, Carmelli G, Sinert R. Pharmacologic cardioversion of recent-onset atrial fibrillation: a systematic review and network meta-analysis. Europace: European Pacing, Arrhythmias, and Cardiac Electrophysiology. 2020; 22(6):854–869 [PubMed: 32176779]

25.: Di Biasi P, Scrofani R, Paje A, Cappiello E, Mangini A, Santoli C. Intravenous amiodarone vs propafenone for atrial fibrillation and flutter after cardiac operation. European Journal of Cardio-Thoracic Surgery. 1995; 9(10):587–591 [PubMed: 8562105]

26.: Ding P, Li L, Zhang H, Chen LJ, Deng XJ, Yuan YQ. Comparison of amiodarone plus irbesartan regimen versus amiodarone alone on maintaining sinus rhythm in rheumatic heart disease patients with persistent atrial fibrillation post valve replacement and cardioversion. Chinese Journal of Cardiovascular Diseases. 2009; 37(6):505–508 [PubMed: 19927630]

27.: Duggan ST, Scott LJ. Intravenous vernakalant: a review of its use in the management of recent-onset atrial fibrillation. Drugs. 2011; 71(2):237–252 [PubMed: 21275448]

28.: Dunning J, Khasati N, Prendergast B. What is the optimal medical treatment for stable cardiac surgical patients who go into atrial fibrillation after their operation? Interactive Cardiovascular and Thoracic Surgery. 2004; 3(1):46–51 [PubMed: 17670174]

29.: Effert S, Merx W. On the treatment of atrial fibrillation following surgery of mitral stenosis. Deutsche Medizinische Wochenschrift. 1967; 92(15):691–695 [PubMed: 6023746]

30.: Fellahi JL, Heringlake M, Knotzer J, Fornier W, Cazenave L, Guarracino F. Landiolol for managing atrial fibrillation in post-cardiac surgery. European Heart Journal Supplements. 2018; 20(Suppl A):A4–A9 [PMC free article: PMC5909770] [PubMed: 30188961]

31.: Fitzgerald RD, Fritsch S, Wislocki W, Oczenski W, Waldenberger FR, Schwarz S. Transvenous, intracardial cardioversion for the treatment of postoperative atrial fibrillation. Annals of Cardiac Anaesthesia. 2008; 11(2):111–115 [PubMed: 18603751]

32.: Frost L, Mortensen PE, Tingleff J, Platou ES, Christiansen EH, Christiansen N. Efficacy and safety of dofetilide, a new class III antiarrhythmic agent, in acute termination of atrial fibrillation or flutter after coronary artery bypass surgery. International Journal of Cardiology. 1997; 58(2):135–140 [PubMed: 9049678]

33.: Gao L, Moodie M. Modelling the lifetime cost-effectiveness of catheter ablation for atrial fibrillation with heart failure. BMJ Open. 2019; 9(9):e031033 [PMC free article: PMC6731807] [PubMed: 31492790]

34.: Gavaghan TP, Koegh AM, Kelly RP, Campbell TJ, Thorburn C, Morgan JJ. Flecainide compared with a combination of digoxin and disopyramide for acute atrial arrhythmias after cardiopulmonary bypass. British Heart Journal. 1988; 60(6):497–501 [PMC free article: PMC1224890] [PubMed: 3146989]

35.: Gillinov AM, Bagiella E, Moskowitz AJ, Raiten JM, Groh MA, Bowdish ME et al. Rate control versus rhythm control for atrial fibrillation after cardiac surgery. New England Journal of Medicine. 2016; 374(20):1911–1921 [PMC free article: PMC4908812] [PubMed: 27043047]

36.: Gong M, Zhang Z, Fragakis N, Korantzopoulos P, Letsas KP, Li G et al. Role of ranolazine in the prevention and treatment of atrial fibrillation: A meta-analysis of randomized clinical trials. Heart Rhythm. 2017; 14(1):3–11 [PubMed: 27746384]

37.: Gray RJ, Conklin CM, Sethna DH, Mandel WJ, Matloff JM. Role of intravenous verapamil in supraventricular tachyarrhythmias after open-heart surgery. American Heart Journal. 1982; 104(4 Pt 1):799–802 [PubMed: 7124593]

38.: Greenberg JW, Lancaster TS, Schuessler RB, Melby SJ. Postoperative atrial fibrillation following cardiac surgery: a persistent complication. European Journal of Cardio-Thoracic Surgery. 2017; 52(4):665–672 [PubMed: 28369234]

39.: Gudbjartsson T, Helgadottir S, Sigurdsson MI, Taha A, Jeppsson A, Christensen TD et al. New-onset postoperative atrial fibrillation after heart surgery. Acta Anaesthesiologica Scandinavica. 2019; [PubMed: 31724159]

40.: Guerra F, Romandini A, Barbarossa A, Belardinelli L, Capucci A. Ranolazine for rhythm control in atrial fibrillation: a systematic review and meta-analysis. International Journal of Cardiology. 2017; 227:284–291 [PubMed: 27839812]

41.: Gun C, Bianco ACM, Freire RB, Ramos RF, Timernan A, Piegas LS. CD-ROM of abstracts from World Cardiology Congress: 1998. Beta-blocker effects on postoperative atrial fibrillation after coronary artery bypass surgery. Journal of American College of Cardiology 1998, Issue 3801.

42.: Ha AC, Mazer CD, Verma S, Yanagawa B, Verma A. Management of postoperative atrial fibrillation after cardiac surgery. Current Opinion in Cardiology. 2016; 31(2):183–190 [PubMed: 26836987]

43.: Habibollahi P, Jam SH, Vahdati SS, Baghi HM, Amiri H. Amiodaron in atrial fibrillation: post coronary artery bypass graft. World Journal of Emergency Medicine. 2016; 7(4):250–254 [PMC free article: PMC5143307] [PubMed: 27942340]

44.: Han QQ, Lang XL, Han L, Lu FL, Zou LJ, Xu ZY. Ibutilide and amiodarone in management of new-onset atrial fibrillation after coronary artery bypass grafting: a comparison of safety and efficacy. Academic Journal of Second Military Medical University. 2010; 31(11):1169–1173

45.: Helber U, Steinhilber B, Fenchel G, Hoffmeister HM. New occurance of post-operative atrial fibrillation after cardio-surgical operations: medical cardioversion verses placebo. Zeitschrift für Kardiologie. 1996; 85(Suppl 2):230

46.: Hilleman DE, Hunter CB, Mohiuddin SM, Maciejewski S. Pharmacological management of atrial fibrillation following cardiac surgery. American Journal of Cardiovascular Drugs. 2005; 5(6):361–369 [PubMed: 16259524]

47.: Hilleman DE, Spinler SA. Conversion of recent-onset atrial fibrillation with intravenous amiodarone: a meta-analysis of randomized controlled trials. Pharmacotherapy. 2002; 22(1):66–74 [PubMed: 11794432]

48.: Hjelms E. Procainamide conversion of acute atrial fibrillation after open-heart surgery compared with digoxin treatment. Scandinavian Journal of Thoracic and Cardiovascular Surgery. 1992; 26(3):193–196 [PubMed: 1287833]

49.: Hogue Jr CW, Hyder ML. Atrial fibrillation after cardiac operation: risks, mechanisms, and treatment. Annals of Thoracic Surgery. 2000; 69(1):300–306 [PubMed: 10654548]

50.: Humphries JO. Unexpected instant death following successful coronary artery bypass graft surgery (and other clinical settings): atrial fibrillation, quinidine, procainamide, et cetera, and instant death. Clinical Cardiology. 1998; 21(10):711–718 [PMC free article: PMC6656189] [PubMed: 9789690]

51.: Hwang MH, Danoviz J, Pacold I, Rad N, Loeb HS, Gunnar RM. Double-blind crossover randomized trial of intravenously administered verapamil. Its use for atrial fibrillation and flutter following open heart surgery. Archives of Internal Medicine. 1984; 144(3):491–494 [PubMed: 6703818]

52.: Iliuta L, Rac-Albu M. Ivabradine versus beta-blockers in patients with conduction abnormalities or left ventricular dysfunction undergoing cardiac surgery. Cardiology and Therapy. 2013; 3(1–2):13–26 [PMC free article: PMC4265234] [PubMed: 25135587]

53.: Iscan S, Eygi B, Besir Y, Yurekli I, Cakir H, Yilik L et al. Inflammation, atrial fibrillation and cardiac surgery: current medical and invasive approaches for the treatment of atrial fibrillation. Current Pharmaceutical Design. 2018; 24(3):310–322 [PubMed: 29384058]

54.: Jannati M. Atrial fibrillation post coronary artery graft surgery: a review of literature. International Journal of General Medicine. 2019; 12:415–420 [PMC free article: PMC6844194] [PubMed: 31807054]

55.: Ji Q, Mei Y, Wang X, Feng J, Wusha D, Cai J et al. Combination of irbesartan and amiodarone to maintain sinus rhythm in patients with persistent atrial fibrillation after rheumatic valve replacement. Circulation Journal. 2010; 74(9):1873–1879 [PubMed: 20668354]

56.: Jie S, Yu C, Xiaohu L, Yanning Q, Zhengnian D, Xiaoyang T. The effect of amiodarone on cardiac rhythm after cardiopulmonary bypass in patients with atrial fibrillation. Journal of Heart Valve Disease. 2010; 19(6):786–788 [PubMed: 21214105]

57.: Johnston BW, Hill R, Duarte R, Chean CS, McAuley DF, Blackwood B et al. Protocol for a systematic review and network meta-analysis of the management of new onset atrial fibrillation in critically unwell adult patients. Systematic Reviews. 2019; 8(1):242 [PMC free article: PMC6816145] [PubMed: 31661022]

58.: Jung W, Meyerfeldt U, Birkemeyer R. Atrial arrhythmias after cardiac surgery in patients with diabetes mellitus. Clinical Research in Cardiology. 2006; 95 (Suppl 1):i88–i97 [PubMed: 16598557]

59.: Kamali A, Sanatkar A, Sharifi M, Moshir E. Evaluation of amiodarone versus metoprolol in treating atrial fibrillation after coronary artery bypass grafting. Interventional Medicine and Applied Science. 2017; 9(2):51–55 [PMC free article: PMC5598126] [PubMed: 28932497]

60.: Kleine P, Dzemali O, Moritz A. Internal cardioversion for treatment of postoperative atrial fibrillation. Surgical Technology International. 2003; 11:197–201 [PubMed: 12931301]

61.: Kolokotroni SM, Toufektzian L, Harling L, Bille A. In patients undergoing lung resection is it safe to administer amiodarone either as prophylaxis or treatment of atrial fibrillation? Interactive Cardiovascular and Thoracic Surgery. 2017; 24(5):783–788 [PubMed: 28453809]

62.: Koplan BA. Adrenergic blockade for management of atrial fibrillation after coronary artery bypass surgery: Is it important to go beyond the beta1-receptor? Heart Rhythm. 2007; 4(9):1175–1176 [PubMed: 17765617]

63.: Kowey PR. Ibutilide for AF termination after cardiac surgery. American Heart Journal. 1997; 134(1):139

64.: Kowey PR, Dorian P, Mitchell LB, Pratt CM, Roy D, Schwartz PJ et al. Vernakalant hydrochloride for the rapid conversion of atrial fibrillation after cardiac surgery: a randomized, double-blind, placebo-controlled trial. Circulation: Arrhythmia and Electrophysiology. 2009; 2(6):652–659 [PubMed: 19948506]

65.: Landymore RW, Howell F. Recurrent atrial arrhythmias following treatment for postoperative atrial fibrillation after coronary bypass operations. European Journal of Cardio-Thoracic Surgery. 1991; 5(8):436–439 [PubMed: 1910852]

66.: Larbuisson R, Venneman I, Stiels B. The efficacy and safety of intravenous propafenone versus intravenous amiodarone in the conversion of atrial fibrillation or flutter after cardiac surgery. Journal of Cardiothoracic and Vascular Anesthesia. 1996; 10(2):229–234 [PubMed: 8850403]

67.: Lee JK, Klein GJ, Krahn AD, Yee R, Zarnke K, Simpson C et al. Rate-control versus conversion strategy in postoperative atrial fibrillation: trial design and pilot study results. Cardiac Electrophysiology Review. 2003; 7(2):178–184 [PubMed: 14618047]

68.: Lee JK, Klein GJ, Krahn AD, Yee R, Zarnke K, Simpson C et al. Rate-control versus conversion strategy in postoperative atrial fibrillation: a prospective, randomized pilot study. American Heart Journal. 2000; 140(6):871–877 [PubMed: 11099990]

69.: Liu TJ, Hsueh CW, Lee WL, Lai HC, Wang KY, Ting CT. Conversion of rheumatic atrial fibrillation by amiodarone after percutaneous balloon mitral commissurotomy. American Journal of Cardiology. 2003; 92(10):1244–1246 [PubMed: 14609612]

70.: Ma W, Guo X, Wang Q, Sun G, Wang J. Systematic review and meta-analysis appraising efficacy and safety of vernakalant for cardioversion of recent-onset atrial fibrillation. Journal of Cardiovascular Pharmacology. 2020; 76(1):32–41 [PubMed: 32251022]

71.: Mahrose R, Elsayed AM, Elshorbagy MS. Bisoprolol versus corticosteroid and bisoprolol combination for prevention of atrial fibrillation after on-pump coronary artery bypass graft surgery. Open Anesthesiology Journal. 2019; 13(1):18–24

72.: Maisel WH, Rawn JD, Stevenson WG. Atrial fibrillation after cardiac surgery. Annals of Internal Medicine. 2001; 135(12):1061–1073 [PubMed: 11747385]

73.: Mangin L, Vinet A, Page P, Glass L. Effects of antiarrhythmic drug therapy on atrioventricular nodal function during atrial fibrillation in humans. Europace: European Pacing, Arrhythmias, and Cardiac Electrophysiology. 2005; 7:(Suppl 2):S71–S82 [PubMed: 16102505]

74.: Mann CJ, Kendall S, Lip GYH. Acute management of atrial fibrillation with acute haemodynamic instability and in the postoperative setting. Heart. 2007; 93(1):45–47 [PMC free article: PMC1861334] [PubMed: 16952970]

75.: Martinez EA, Bass EB, Zimetbaum P, American College of Chest P. Pharmacologic control of rhythm: American College of Chest Physicians guidelines for the prevention and management of postoperative atrial fibrillation after cardiac surgery. Chest. 2005; 128(Suppl 2):48S–55S [PubMed: 16167665]

76.: Martinez EA, Epstein AE, Bass EB, American College of Chest P. Pharmacologic control of ventricular rate: American College of Chest Physicians guidelines for the prevention and management of postoperative atrial fibrillation after cardiac surgery. Chest. 2005; 128(Suppl 2):56S–60S [PubMed: 16167666]

77.: McAlister HF, Luke RA, Smith WM. Amiodarone versus quinidine reversion of atrial fibrillation post-cardiac surgery. Australian and New Zealand Journal of Medicine. 1986; 16:588

78.: McAlister HF, Luke RA, Whitlock RM, Smith WM. Intravenous amiodarone bolus versus oral quinidine for atrial flutter and fibrillation after cardiac operations. The Journal of thoracic and cardiovascular surgery. 1990; 99(5):911–918 [PubMed: 2184287]

79.: McKeown PP, American College of Chest P. Introduction: American College of Chest Physicians guidelines for the prevention and management of postoperative atrial fibrillation after cardiac surgery. Chest. 2005; 128(Suppl 2):6S–8S [PubMed: 16167658]

80.: McMurry SA, Hogue Jr CW. Atrial fibrillation and cardiac surgery. Current Opinion in Anaesthesiology. 2004; 17(1):63–70 [PubMed: 17021530]

81.: Mooss AN, Wurdeman RL, Mohiuddin SM, Reyes AP, Sugimoto JT, Scott W et al. Esmolol versus diltiazem in the treatment of postoperative atrial fibrillation/atrial flutter after open heart surgery. American Heart Journal. 2000; 140(1):176–180 [PubMed: 10874282]

82.: Morttada A, Ramzy A, Ayoub A, Khaled S, Samir S, Zaki H. Effects of atrial fibrillation cardioversion after percutaneous mitral balloon valvuloplasty on echocardiographic left and right atrial functions. Journal of Atrial Fibrillation. 2016; 8(6):48–50 [PMC free article: PMC5089465] [PubMed: 27909490]

83.: National Institute for Health and Care Excellence. Developing NICE guidelines: the manual [Updated October 2018]. London. National Institute for Health and Care Excellence, 2014. Available from: https://www.nice.org.uk/process/pmg20/chapter/introduction-and-overview [PubMed: 26677490]

84.: Nemati MH, Astaneh B. Amiodarone versus propafenone to treat atrial fibrillation after coronary artery bypass grafting: a randomized double blind controlled trial. The Korean Journal of Thoracic and Cardiovascular Surgery. 2016; 49(3):177–184 [PMC free article: PMC4900860] [PubMed: 27298795]

85.: O’Bryan LJ, Redfern OC, Bedford J, Petrinic T, Young JD, Watkinson PJ. Managing new-onset atrial fibrillation in critically ill patients: a systematic narrative review. BMJ Open. 2020; 10(3):e034774 [PMC free article: PMC7202704] [PubMed: 32209631]

86.: Ozaydin M, Erdogan D, Yucel H, Peker O, Icli A, Akcay S et al. N-acetyl cysteine for the conversion of atrial fibrillation into sinus rhythm after cardiac surgery: a prospective, randomized, double-blind, placebo-controlled pilot study. International Journal of Cardiology. 2013; 165(3):580–583 [PubMed: 23041016]

87.: Patel D, Gillinov MA, Natale A. Atrial fibrillation after cardiac surgery: where are we now? Indian Pacing and Electrophysiology Journal. 2008; 8(4):281–291 [PMC free article: PMC2572025] [PubMed: 18982137]

88.: Plumb VJ, Karp RB, Kouchoukos NT, Zorn GL, Jr., James TN, Waldo AL. Verapamil therapy of atrial fibrillation and atrial flutter following cardiac operation. The Journal of thoracic and cardiovascular surgery. 1982; 83(4):590–596 [PubMed: 7038316]

89.: Qian YJ, Xiao XJ, Yuan HS, Tang H, Shao HZ, Wei DM. Combination pharmacological cardioversion of permanent atrial fibrillation in post-prosthetic mitral valve replacement outpatients: a novel approach for the treatment of atrial fibrillation. Journal of International Medical Research. 2008; 36(3):537–543 [PubMed: 18534136]

90.: Rho RW, Bridges CR, Kocovic D. Management of postoperative arrhythmias. Seminars in Thoracic and Cardiovascular Surgery. 2000; 12(4):349–361 [PubMed: 11154730]

91.: Rosenberg K. Rate and rhythm control are equally effective in treating postoperative atrial fibrillation. American Journal of Nursing. 2016; 116(8):53–54 [PubMed: 27466928]

92.: Sai C, Li J, Ruiyan M, Yingbin X. Atorvastatin prevents postoperative atrial fibrillation in patients undergoing cardiac surgery. Hellenic Journal of Cardiology. 2019; 60(1):40–47 [PubMed: 29307691]

93.: Sakamoto A, Kitakaze M, Takamoto S, Namiki A, Kasanuki H, Hosoda S et al. Landiolol, an ultra-short-acting beta1-blocker, more effectively terminates atrial fibrillation than diltiazem after open heart surgery: prospective, multicenter, randomized, open-label study (JL-KNIGHT study). Circulation Journal. 2012; 76(5):1097–1101 [PubMed: 22361918]

94.: Savelieva I, Graydon R, Camm AJ. Pharmacological cardioversion of atrial fibrillation with vernakalant: evidence in support of the ESC Guidelines. Europace: European Pacing, Arrhythmias, and Cardiac Electrophysiology. 2014; 16(2):162–173 [PubMed: 24108230]

95.: Schleifer JW, Ramakrishna H. Perioperative atrial fibrillation: a systematic review of evolving therapeutic options in pharmacologic and procedural management. Annals of Cardiac Anaesthesia. 2015; 18(4):565–570 [PMC free article: PMC4881665] [PubMed: 26440243]

96.: Schwartz M, Michelson EL, Sawin HS, MacVaugh H, 3rd. Esmolol: safety and efficacy in postoperative cardiothoracic patients with supraventricular tachyarrhythmias. Chest. 1988; 93(4):705–711 [PubMed: 2894920]

97.: Selvaraj T, Kiran U, Das S, Chauhan S, Sahu B, Gharde P. Effect of single intraoperative dose of amiodarone in patients with rheumatic valvular heart disease and atrial fibrillation undergoing valve replacement surgery. Annals of Cardiac Anaesthesia. 2009; 12(1):10–16 [PubMed: 19136749]

98.: Shantsila E, Watson T, Lip GYH. Atrial fibrillation post-cardiac surgery: changing perspectives. Current Medical Research and Opinion. 2006; 22(8):1437–1441 [PubMed: 16870069]

99.: Shen SL, Zhao YC. A comparative study on the efficacy and safety of intravenous esmolol, amiodarone and diltiazem for controlling rapid ventricular rate of patients with atrial fibrillation during anesthesia period. Chinese Journal of Cardiovascular Diseases. 2010; 38(11):989–992 [PubMed: 21215227]

100.: Simopoulos V, Hevas A, Hatziefthimiou A, Dipla K, Skoularigis I, Tsilimingas N et al. Amiodarone plus ranolazine for conversion of post-cardiac surgery atrial fibrillation: enhanced effectiveness in reduced versus preserved ejection fraction patients. Cardiovascular Drugs and Therapy. 2018; 32(6):559–565 [PubMed: 30255400]

101.: Simopoulos V, Tagarakis GI, Daskalopoulou SS, Daskalopoulos ME, Lenos A, Chryssagis K et al. Ranolazine enhances the antiarrhythmic activity of amiodarone by accelerating conversion of new-onset atrial fibrillation after cardiac surgery. Angiology. 2014; 65(4):294–297 [PubMed: 23427280]

102.: Somberg J, Molnar J. Sotalol versus amiodarone in treatment of atrial fibrillation. Journal of Atrial Fibrillation. 2016; 8(5):1359 [PMC free article: PMC5089490] [PubMed: 27909477]

103.: Soucier R, Silverman D, Abordo M, Jaagosild P, Abiose A, Madhusoodanan KP et al. Propafenone versus ibutilide for post operative atrial fibrillation following cardiac surgery: neither strategy improves outcomes compared to rate control alone (the PIPAF study). Medical Science Monitor. 2003; 9(3):19–23 [PubMed: 12640352]

104.: Stiell IG, Sivilotti MLA, Taljaard M, Birnie D, Vadeboncoeur A, Hohl CM et al. Electrical versus pharmacological cardioversion for emergency department patients with acute atrial fibrillation (RAFF2): a partial factorial randomised trial. Lancet. 2020; 395(10221):339–349 [PubMed: 32007169]

105.: Szyszka A, Wachowiak-Baszyńska H, Paluszkiewicz L, Skorupski W, Brocki Z. Preventive pharmacological treatment after cardiac electroversion for atrial fibrillation following surgical treatment of heart defects. Kardiologia Polska. 1992; 37(8):74–78 [PubMed: 1434328]

106.: Taenaka N, Kikawa S. The effectiveness and safety of landiolol hydrochloride, an ultra-short-acting beta1-blocker, in postoperative patients with supraventricular tachyarrhythmias: a multicenter, randomized, double-blind, placebo-controlled study. American Journal of Cardiovascular Drugs. 2013; 13(5):353–364 [PMC free article: PMC3781301] [PubMed: 23818039]

107.: Tisdale JE, Padhi ID, Goldberg AD, Silverman NA, Webb CR, Higgins RS et al. A randomized, double-blind comparison of intravenous diltiazem and digoxin for atrial fibrillation after coronary artery bypass surgery. American Heart Journal. 1998; 135(5 Pt 1):739–747 [PubMed: 9588402]

108.: Tsu LV, Lee S. Use of ranolazine in the prevention and treatment of postoperative atrial fibrillation in patients undergoing cardiac surgery. Annals of Pharmacotherapy. 2014; 48(5):633–637 [PubMed: 24523397]

109.: VanderLugt JT, Mattioni T, Denker S, Torchiana D, Ahern T, Wakefield LK et al. Efficacy and safety of ibutilide fumarate for the conversion of atrial arrhythmias after cardiac surgery. Circulation. 1999; 100(4):369–375 [PubMed: 10421596]

110.: Vilvanathan VK, Srinivas Prabhavathi Bhat BC, Nanjappa MC, Pandian B, Bagi V, Kasturi S et al. A randomized placebo-controlled trial with amiodarone for persistent atrial fibrillation in rheumatic mitral stenosis after successful balloon mitral valvuloplasty. Indian Heart Journal. 2016; 68(5):671–677 [PMC free article: PMC5079138] [PubMed: 27773406]

111.: Wafa SS, Ward DE, Parker DJ, Camm AJ. Efficacy of flecainide acetate for atrial arrhythmias following coronary artery bypass grafting. American Journal of Cardiology. 1989; 63(15):1058–1064 [PubMed: 2495710]

112.: White CM, Nguyen E. Novel use of ranolazine as an antiarrhythmic agent in atrial fibrillation. Annals of Pharmacotherapy. 2017; 51(3):245–252 [PubMed: 27758968]

113.: Yilmaz AT, Demirkilic U, Arslan M, Kurulay E, Ozal E, Tatar H et al. Long-term prevention of atrial fibrillation after coronary artery bypass surgery: comparison of quinidine, verapamil, and amiodarone in maintaining sinus rhythm. Journal of Cardiac Surgery. 1996; 11(1):61–64 [PubMed: 8775337]

Appendices

Appendix A. Review protocols

Table 21. Review protocol: Treatment of atrial fibrillation after cardiothoracic surgery (PDF, 279K)

Table 22. Health economic review protocol (PDF, 165K)

Appendix B. Literature search strategies

This literature search strategy was used for the following reviews:

What is the most clinical and cost-effective treatment strategy (rate or rhythm control or no treatment) for people with atrial fibrillation after cardiothoracic surgery?

The literature searches for this review are detailed below and complied with the methodology outlined in Developing NICE guidelines: the manual.⁸³

For more information, please see the Methods Report published as part of the accompanying documents for this guideline.

B.1. Clinical search literature search strategy (PDF, 321K)

B.2. Health Economics literature search strategy (PDF, 239K)

Appendix H. Health economic evidence tables

None.

Appendix I. Excluded studies

I.1. Excluded clinical studies (PDF, 158K)

I.2. Excluded health economic studies (PDF, 98K)

Final

Intervention evidence review

Developed by the National Guideline Centre, Royal College of Physicians

Disclaimer: The recommendations in this guideline represent the view of NICE, arrived at after careful consideration of the evidence available. When exercising their judgement, professionals are expected to take this guideline fully into account, alongside the individual needs, preferences and values of their patients or service users. The recommendations in this guideline are not mandatory and the guideline does not override the responsibility of healthcare professionals to make decisions appropriate to the circumstances of the individual patient, in consultation with the patientand, where appropriate,their carer or guardian.

Local commissioners andproviders have a responsibility to enable the guideline to be applied when individual health professionals and theirpatients or service users wish to use it. They should do so in the context of local and national priorities for funding and developing services, and in light of their duties to have due regard to the need to eliminate unlawful discrimination, to advance equality of opportunity and to reduce health inequalities. Nothing in this guideline should be interpreted in a way that would be inconsistent with compliance with those duties.

NICE guidelines cover health and care in England. Decisions on how they apply in other UK countries are made by ministers in the Welsh Government, Scottish Government, and Northern Ireland Executive. All NICE guidance is subject to regular review and may be updated or withdrawn.

Bookshelf ID: NBK571341PMID: 34165940

Table 1PICO characteristics of review question

Population	People aged over 18 who have had cardiothoracic surgery and who have post-operative AF (stratified by pre-existing AF vs. no pre-existing AF)
Interventions	Rate control strategies (lists below are not exhaustive): Beta blockers -for example, bisoprolol, acebutolol, metoprolol, nadolol, pindolol, betaspace, propranolol, esmalol Ca2+ channel blockers – for example, diltiazem hydrochloride, verapamil Digoxin Amiodarone* Rhythm control strategies (lists below are not exhaustive): Na+ channel blockers – such as procainamide, disopyramide, quinidine sulphate, flecainide, propafenone K+ channel blockers – such as amiodarone, dronedarone, ibutilide, sotalol DC cardioversion amiodarone may be used for rate or rhythm control.
Comparisons	Placebo No treatment To each other (between classes of intervention – i.e. beta blockers vs Ca channel blockers, or digoxin versus DC cardioversion). RCTs where individuals in the intervention group may be prescribed different rate or rhythm drugs are allowed. No comparisons within classes will be included (i.e. bisoprolol versus pindolol, or procainamide versus flecainide)
Outcomes	Critical health-related quality of life mortality stroke or thromboembolic complications Need for rescue DC cardioversion Rehospitalisation (all cause) Rehospitalisation for AF Achievement of sinus rhythm Adverse events Important freedom from anticoagulation freedom from AAD use Hospital length of stay ICU length of stay
Study design	Randomised controlled trials and systematic reviews of RCTs

Table 2Summary of studies included in the evidence review

Study

Intervention and comparison

Population

Outcomes

Comments

Chen 2013¹³

RCT

N=115

Conducted in China

DC cardioversion: Initial 200mg amiodarone three times daily for four days. Electrical defibrillation was performed on fifth day. Once unconscious, patients received direct-current synchronized electrical cardioversion using an initial energy level of 200 J. If the first shock failed, two 300 J shocks were given.

Electrical treatment was stopped if sinus rhythm was not restored within three shocks.

After reversion to normal rhythm, 200 mg amiodarone was taken daily for 30 days

K+ blockers + captopril + simvastatin: Oral amiodarone with captoptri land simvastatin.

Three months combination therapy with oral amiodarone at a dose of 600 mg/day for 3 days, 400 mg/day for the following 3 days and then 200 mg/day, oral captopril at a dose of 12.5 mg twice daily before food and oral simvastatin at a dose of 15 mg/day at night.

Pre-existing AF stratum: those with pre-existing permanent AF prior to mitral valve replacement operation.

Inclusion criteria:

Age >18 years with permanent AF
Undergone prosthetic mitral valve replacement with or without aortic valve replacement
Cardiothoracic ratio ≤0.5 on cardiac anterioposterior X-radiography and a left atrial diameter ≤50 mm on Doppler ultrasound for ≥6 months post-surgery

Exclusion criteria:

New York Heart Association heart failure class IV
History of sick sinus syndrome or second-or third-degree atrioventricular block
Severe hepatic and/or renal dysfunction
Hyperthyroidism
Contraindications to treatment with amiodarone

Population characteristics:

Mean (SD) duration of AF: 54.2 (25.9) vs. 53.5 (25.4) months
Mean (SD) LVEF: 58.6 (6.5) vs. 58.2 (6.2)%

Mortality
Rehospitalisation for AF
Achievement of sinus rhythm
Adverse events

All patients received standard long-term anticoagulant therapy with warfarin and/or digoxin following surgery.

Chen 2019¹²

RCT

N=84

Conducted in China

K+ blockers(amiodarone):On the day of the operation, a micro infusion pump administered 600 mg amiodarone at 50 mg/h for 12 h. On postoperative day 1, amiodarone taken orally following recovery of diet 3 times daily. A week later, patients took amiodarone 2 times daily. Another week later, the dose was reduced to once daily. Treatment course was 1 month. Also received routine medical treatment as described in the control group.

Routine medical treatment: Routine treatment consisted of oral administration of drugs for diuresis, anticoagulation and routine application of antibiotics.

No pre-existing AF stratum: states AF developed following valve replacement

Inclusion criteria:

Elective valvular replacement
Rheumatic heart disease with continuous atrial fibrillation
Cardiac function no higher than grade III
Satisfied application indication for amiodarone
Normal levels of electrolytes and acidity and alkalinity
Heart rate <70 bpm

Exclusion criteria:

Presence of other types of arrhythmia

Population characteristics:

Mean (SD) duration of AF: 30 (14.93) vs. 31.06 (15.02) months
Mean (SD) left ventricular ejection fraction: 46.7 (4.32) vs. 47.5 (4.03)%
Mean (SD) New York Heart Association score: 2.50 (0.51) vs. 2.58 (0.50)

Achievement of sinus rhythm
Hospital length of stay
ICU length of stay

Cochrane 1994¹⁵

RCT

N=30

Conducted in Australia

K+ blockers: Intravenous amiodarone. Loading dose of 5 mg/kg (max. 400 mg) in 100 ml of 5% dextrose infused intravenously over 30 min.

At 30 min after loading dose complete, infusion of 25 mg/h initiated. Infusion increased to 40 mg/h if ventricular rate still exceeded 120 beats/min after 6 h. Treatment continued for 24 h after reversion to sinus rhythm.

If no reversion following 24 h of intravenous amiodarone infusion, digoxin was added at half the dose used in the digoxin treatment group

Digoxin: Intravenous digoxin. Loading dose of 1 mg given intravenously over 9 h as follows: 0.5 mg over 30 min at start of treatment, followed by 0.25 mg after 2 h and 0.125 mg after 5 h and 9 h.

Oral maintenance therapy started within 12 h at a dose suitable for body weight and renal function.

If reversion did not occur during 24 h treatment period amiodarone was added at dose described for amiodarone group and digoxin continued at half the previous dose

No pre-existing AF stratum: AF prior to surgery was an exclusion criterion

Inclusion criteria:

People that developed AF persisting for >20 min with systolic blood pressure of ≥85 mmHg without inotropic support, while recovering from open heart surgery

Exclusion criteria:

AF prior to surgery
Poor ventricular contractility on preoperative left venticulogram
Postoperative administration of beta-blockers

Population characteristics:

Operation type: coronary artery bypass grafting(73 vs. 67%), aortic valve replacement(20 vs. 20%), mitral valvotomy(7 vs. 0%)or combined procedures(0 vs. 13%)
Pre-operative beta-blockade: 47 vs. 53%

Need for rescue DC cardioversion
Achievement of sinus rhythm
Adverse events

Fitzgerald 2008³¹

RCT

N=18

Conducted in Austria

DC cardioversion

ALERT catheter for intracardiac conversion.

Sedation with midazolam (3–5 mg intravenously) and intracardiac cardioversion.

ALERT system provides temporary pacing, sensing and delivery of stimuli for internal cardioversion.

First shock with 3 J. Following each shock, 12-lead ECG was obtained to assess sinus rhythm. If rhythm was not converted, shock energy was increased by increments of 3 J to a maximum of 15 J.

If no response after 15 J shock, the patient was classified as being not responsive. When awake and haemodynamically stable patients were returned to the ward. Non-responders received treatment according to the preference of the doctor in charge and were excluded from further evaluation

K+ blockers

Standard pulmonary artery catheter for delivery of intravenous amiodarone. Bolus dose of 250 mg followed by continuous infusion of 0.6 mg/kg/h. Duration unclear.

No pre-existing AF stratum: chronic AF and conduction disorders were exclusion criteria

Inclusion criteria:

Undergoing coronary artery bypass grafting or valve surgery
Postoperative AF development

Exclusion criteria:

Chronic atrial fibrillation
Conduction disorders
Patients with pacemakers

Population characteristics:

Type of surgery: CABG (44 vs. 56%), aortic valve (22 vs. 22%), mitral valve (11 vs. 11%), CABG + valve (22 vs. 11%)
NYHA class: I (11 vs. 11%), II (56 vs. 67%), III (22 vs. 11%), IV (11 vs. 11%)
Median (range) ejection fraction: 53 (31–74) vs. 59 (36–68)%

Need for rescue DC cardioversion
Achievement of sinus rhythm

In both groups: Patients allowed treatment with intravenous digoxin if tachycardia led to haemodynamic instability at any time prior to the intervention. If this treatment was not successful in establishing stability and further treatment was required, patients were excluded from further evaluation

Gillinov 2016³⁵

RCT

N=523

Conducted in Canada and USA

Mixed rate control

Received medications to slow heart rate with aim of achieving a resting heart rate <100 beats/min.

Patients in whom sinus rhythm was not restored after rate control could be switched to rhythm control if provider thought necessary to improve haemodynamic status or alleviate symptoms.

Duration 60 days. No further details on the interventions given.

K+ blocker with/without rate control agent

Amiodarone with or without rate-slowing agent.

If atrial fibrillation persisted for 24–48 h after randomisation, direct current cardioversion was recommended.

Recommended dose of amiodarone was 3 g of oral amiodarone before hospital discharge with a maintenance dose of 200 mg/day or less if direct current cardioversion was successful.

It was recommended that the use of amiodarone be extended for 60 days, but discontinuation was allowed for amiodarone-related adverse events, such as bradycardia, corrected QT interval >480 msec or neuropathy

No pre-existing AF: History of AF was an exclusion criterion

Inclusion criteria:

Haemodynamically stable adults
Undergone elective surgery to treat coronary artery disease or heart valve disease
Postoperative atrial fibrillation persisting for >60 min or recurrent episodes of atrial fibrillation during index hospitalisation

Exclusion criteria:

History of atrial fibrillation

Patient characteristics:

Diabetes: 31.3 vs. 30.3%
Heart failure: 13.4 vs. 12.6%
Hypertension: 73.7 vs. 75.9%
Previous myocardial infarction: 19.1 vs. 18.4%
Previous stroke: 6.5 vs. 5.7%
Medication: ACE inhibitor (34 vs. 32.2%), ARB (19.5 vs. 18%), beta-blocker (61.8 vs. 55.6%), calcium channel blocker (19.8 vs. 22.2%), diuretic (30.2 vs. 31%), nitrate (22.9 vs. 21.1%)
Index operation: CABG only (42.7 vs. 38.3%), valve repair only (14.9 vs. 16.5%), CABG + valve repair (3.8 vs. 2.7%), valve replacement only (22.9 vs. 25.3%), CABG + valve replacement (15.6 vs. 17.2%)

Rehospitalisation, all-cause
Mortality
Stroke or thromboembolic complications
Need for rescue DC cardioversion
Rehospitalisation for AF
Achievement of sinus rhythm
Adverse events
Hospital length of stay
Freedom from anticoagulation

For both groups:

If patients remained in atrial fibrillation or had recurrent atrial fibrillation 48 hafter randomization, anticoagulation with warfarin (target international normalized ratio, 2 to 3) was recommended, and bridging with low-molecular-weight heparin was allowed. Anticoagulation was recommended to be continued for 60 days, unless complications occurred

Gray 1982³⁷

RCT

N=22

Conducted in USA

Calcium channel blockers

Intravenous verapamil. Low-dose verapamil (0.075 mg/kg body weight, up to maximum of 10 mg) administered as bolus intravenous injection over 1 min.

Placebo

Intravenous placebo. Volume of placebo similar to that used in verapamil administration was administered as bolus intravenous injection over 1 min.

Mixed/unclear stratum: 18.1% had history of atrial tachyarrhythmias Inclusion criteria:

Development of supraventricular tachyarrhythmias (atrial fibrillation, atrial flutter or atrial tachycardias) following open heart surgery.

Exclusion criteria:

Age >74 years or <21 years
Evidence of renal or hepatic failure
Received propanolol24 h previously

Patient characteristics (not given separately for each group):

Arrhythmia type: Atrial fibrillation, 81.8%; atrial flutter, 18.2%
Type of surgery: CABG, 86.4%; double valve (aortic and mitral) replacement, 9.09%; mitral commissurotomy and aortic valve replacement, 4.56%
Digoxin had been given in 20 patients (mean dose 0.5 mg) within the 24 h prior to verapamil administration (when verapamil used either as first drug or when used as second drug if placebo failed)

Adverse events

Those that did not achieve positive response with first drug were switched over to the other drug, which may affect outcomes. Analysed as randomised in review unless stated otherwise.

For both groups:

A positive response was observed if the patient converted to sinus rhythm or the heart rate decreased below 100 bpm (120 bpm in digoxin-treated patients).
If no positive response was seen within 10 min, a higher dose of the initial intervention was administered consisting of 0.15 mg/kg verapamil or placebo equivalent.
If no response was seen after 15 min, drug B (placebo or verapamil) was administered first in low dose and then in high dose as described for each intervention.

Digoxin had been given in 20 patients (mean dose 0.5 mg) within the 24 h prior to verapamil administration (when verapamil used either as first drug or when used as second drug if placebo failed)

Hwang 1984⁵¹

RCT

N=14

Conducted in USA

Calcium channel blockers

Intravenous verapamil. First dose of0.075 mg/kg given intravenously over 1 min.

If therapeutic end point was not achieved after 15 min, administration of verapamil was repeated at a dose of 0.15 mg/kg, up to a maximum of 10 mg.

After a further 30 min, if therapeutic end point not achieved, the second drug (placebo) was administered in a similar fashion

Placebo

Intravenous placebo. First dose of 0.075 mg/kg placebo given intravenously over 1 min.

If therapeutic end point was not achieved after 15 min, administration of placebo was repeated at a dose of 0.15 mg/kg, up to a maximum of 10 mg.

After a further 30 min, if therapeutic end point not achieved, the second drug (verapamil) was administered in a similar fashion

No pre-existing AF stratum: being in normal sinus rhythm prior to and immediately after surgery was inclusion criterion.

Inclusion criteria:

Presence of sinus rhythm prior to and immediately following surgery
Development of supraventricular tachycardia with a ventricular response >120 beats/min during postoperative period and persisting for at least 1 h

Exclusion criteria:

Preoperative left ventricular ejection fraction <45% or clinical postoperative cardiac failure
Hypotension (<systolic pressure below 90 mmHg)
Notable valvular heart disease
Impaired atrioventricular conduction or evidence of depressed sinus node automaticity
Impaired hepatic or renal function
Administration of beta-blocking drugs or disopyramide within previous 48 h

Patient characteristics (not given separately for each group):

Arrhythmia type: atrial fibrillation, 78.6%; atrial flutter, 21.4%
Mean (range) preoperative LVEF: 62 (49–74)%
Type of surgery: aortocoronary bypass, 85.7%; aortocoronary bypass + thymectomy, 7.1%; atrial septal defect closure, 7.1%

Achievement of sinus rhythm
Adverse events

Those that did not achieve positive response with first drug were switched over to the other drug, which may affect outcomes. Analysed as randomised in review unless stated otherwise.

Kowey 2009⁶⁴

K+ blockers

Infusion of 3.0 mg/kg vernakalant over 10 min followed by 15 min observation period. Second infusion of 2.0 mg/kg vernakalant performed over 10 min if did not convert to sinus rhythm.

Placebo

Infusion of 3.0 mg/kg placebo (normal saline) over 10 min followed by 15 min observation period. Second infusion of 2.0 mg/kg placebo performed over 10 min if did not convert to sinus rhythm.

No pre-existing AF stratum: being in normal sinus rhythm prior to and immediately after surgery was inclusion criterion.

Inclusion criteria:

≥18 years old with sustained AF or flutter occurring between 24 h and 7 days following surgery.
Haemodynamically stable
Sinus rhythm before and after surgery
Weight between 46 and 136 kg

Exclusion criteria:

Pregnancy or nursing
Uncorrected QT interval >500 ms
Ventricular response rate to AF <45 bpm
QRS interval >140 ms without pacemaker
Second or third-degree AV block
History of torsadesde pointes
Unstable class IV congestive heart failure, serious hepatic or renal disease, or end-stage disease states
Reversible cause of AF (e.g. hyperthyroidism or pulmonary embolism)
Uncorrected electrolyte imbalance
Digoxin toxicity
Received another investigational drug or IV vernakalant in past 30 days, received oral amiodarone with past 3 months, received IV amiodarone within past 24 h, or received class I or III antiarrhythmic drugs following cardiac surgery

Mortality
Achievement of sinus rhythm
Adverse events (serious and treatment-emergent)

Infusion was discontinued if one of various adverse effects were observed, such as uncorrected QT interval ≥550 ms or prolongation of the uncorrected QT interval >25%

Lee 2000⁶⁸ and Lee 2003⁶⁷

RCT

N=50

Conducted in Canada

Mixed rhythm control with/without electrical cardioversion

Antiarrhythmic drug therapy with or without electrical cardioversion. Aimed at restoration of sinus rhythm within 48 h.

Preferred initial treatment was with sotalol or propafenone, taking into consideration left ventricular function, history of coronary artery disease and contraindications to beta-blockers.

Sotalol prescribed at dose of 120–360 mg/day; amiodarone at 200 mg/day after a loading dose of 1200–1600 mg for 4–5 days; and propafenone at dose of 300–900 mg/day. Procainamide given as intravenous load of 500–1000 mg followed by continuous infusion of 1–4 mg/h or 2–3 g/day in divided oral doses.

If sinus rhythm was not achieved within 48 h, patients were electrically cardioverted

Rate control therapy used in rate control group were permitted if clinically indicated (e.g. ventricular rates ≥110 per min) and in patients who received propafenone because of potential for 1:1 AV conduction during atrial fibrillation.

Mixed rate control

Preferred initial treatment was IVdiltiazem for those requiring IVagent on basis of symptom severity and beta-blockers in those treated with oral agents.

IV diltiazem administered as initial bolus of 5–20 mg followed by continuous infusion of 5–15 mg/h.
Oral diltiazem given at 120–360 mg/day and verapamil given orally in similar fashion.
Metoprolol given at dose of 25–100 mg/day in 2 divided doses
Atenolol given at dose of 25–100 mg/day
Propanolol given at a dose of 30–120 mg in 3 divided doses
Esmolol given at 0.05 mg/kg per minute intravenous loading followed by maintenance dose of 0.05–0.2 mg/kg per minute.
Digoxin loading administered either intravenously or orally:
- Oral loading dose of 0.25–0.5 mg digoxin was given followed by 0.25 mg every 4–6 hours until a loading dose of 1 mg had been given.
- Intravenous digoxin administered in similar fashion.
- Daily maintenance dose of 0.25 mg was administered thereafter for digoxin

No pre-existing AF stratum: those with history of paroxysmal AF or conduction disturbances at randomisation excluded

Inclusion criteria:

Development of atrial fibrillation lasting at least 1 h following heart surgery
Ability to give informed consent
18 years of age or above

Exclusion criteria:

History of paroxysmal atrial fibrillation
Received antiarrhythmic therapy within 5 half-lives of the time of randomisation
Had beta-blockers withdrawn after surgery
Were in cardiogenic shock
Creatinine level of >200 μg/mmol
Serum aspartate aminotransferase or alanine aminotransferase concentrations 4 times the upper limit of normal;
Conduction disturbances before randomisation
Contraindications to anticoagulation

Patient characteristics:

Mean (SD) LVEF: 49 (1) vs. 47(11)%
Preoperative beta-blockers: 63 vs. 61%
Preoperative calcium channel blockers: 63 vs. 36%
Valvular surgery: 30 vs. 30%
Diabetes: 19 vs. 23%
Hypertension: 48 vs. 52%

Mortality
Achievement of sinus rhythm
Hospital length of stay

For both groups:

Intravenous heparin and oral warfarin started within 24 h after randomisation.

Dose of heparin was titrated to maintain a partial thromboplastin time between 80 and 100 seconds.
Warfarin doses adjusted to obtain an INR between 2 and 3.
Anticoagulation continued until end of study

Nemati 2016⁸⁴

RCT

N=122

Conducted in Iran

Na+ blockers

Oral propafenone. 600 mg loading dose and 150 mg every 8 h for 10 days after onset of atrial fibrillation.

If AF did not resolve after this first dosing strategy, it could be repeated or switched to amiodarone

K+ blockers

Intravenous amiodarone. 300 mg intravenous loading dose followed by continuous intravenous infusion of 600 mg over 12–24 h after the onset of atrial fibrillation.

If AF did not resolve after this first dosing strategy, it could be repeated or switched to propafenone

No pre-existing AF stratum: History of AF within previous 6 months an exclusion criterion. <10% with AF history before this period.

Inclusion criteria:

Development of postoperative atrial fibrillation (continuous AF for at least 30 min or AF requiring treatment for symptoms or hemodynamic compromise) following elective coronary artery bypass grafting

Exclusion criteria:

Patients that underwent concomitant cardiac operations at same time as coronary artery bypass grafting
Bradycardia (<50 beats/min in resting position)
Patients with > type I second-degree heart block
Symptomatic sick sinus syndrome without a pacemaker
Taking class I or III antiarrhythmic medications
History of AF within past 6 months
Sensitivity to propafenone
Cardiogenic shock
Ejection fraction <30%
Marked hypotension (systolic blood pressure <90 mmHg)
Electrolyte imbalances

Patient characteristics:

All underwent CABG surgery
Hypertension: 70.9 vs. 77.6%
Hyperlipidaemia: 69.1 vs. 67.2%
Diabetes mellitus: 50.9 vs. 49.3%
Congestive heart failure: 0 vs. 3.1%
Previous atrial fibrillation: 9.1 vs. 3.2%
Drugs: beta-blockers (87.3 vs. 80.6%), calcium channel blockers (9.1 vs. 12.5%), ACE inhibitor (34.5 vs. 25.8%)

Need for rescue DC cardioversion
Achievement of sinus rhythm
Adverse events

If the treatment strategy patients were assigned to did not resolve the AF after 10 days, treatment with drug could be repeated or switched to other drug, which could affect outcomes. Analysed as randomised in the review unless otherwise stated.

Qian 2008⁸⁹

RCT

N=99

Conducted in China

Mixed rate control

Control of ventricular rate using digoxin and diltiazem, either alone or in combination.

Doses of drugs not stated. Appears to be oral dosing based on length of the study but not explicitly stated

Duration, 12 months.

K+ blockers + captopril +simvastatin

Pharmacological cardioversion with low-dose oral amiodarone (2 mg/kg), captopril (0.25 mg/kg) and simvastatin (0.3 mg/kg) administered daily.

Heart rate was maintained at 60–80 beats/min under quiescent conditions.

If needed, digoxin and/or diltiazem were also administered in these patients

Duration, 12 months.

Pre-existing AF: duration of AF longer than time since surgery, indicating AF present prior to surgery

Inclusion criteria:

>18 years of age
Permanent atrial fibrillation for at least 6 months following prosthetic mitral valve replacement

Exclusion criteria:

Moderate or severe tricuspid regurgitation
NYHA heart failure class IV
History of sick sinus syndrome or second-or third-degree atrioventricular block
Significant thyroid, pulmonary or hepatic disease
Contraindications to treatment with amiodarone
Significant impairment of renal function
Pregnancy or females shortly intending to become pregnant
Any other medical condition that in the opinion of the investigators could make the patient inappropriate for the study

Patient characteristics:

All received mitral valve surgery
Mean (SD) duration of AF: 35.7 (16.1) vs. 36.3 (17.5) months
Mean (SD) LVEF: 46 (13.1) vs. 45.7 (12.1)%

Achievement of sinus rhythm
Adverse events

Simonpoulos 2014¹⁰¹

RCT

N=41

Conducted in Greece

K+ blockers

IVamiodarone. 300 mg in 30 min followed by 750 mg in 24 h.

After conversion to sinus rhythm the amiodarone infusion was stopped but received amiodarone orally at a dose of 200 mg twice daily for a week and 200 mg once daily for the second week, or according to their cardiologist’s advice following discharge

K+ blockers + ranolazine

IV amiodarone at 300 mg in 30 min followed by 750 mg in 24 h. Oral ranolazine regimen consisted of 500 mg loading dose followed by 375 mg 6 hours later and then 375 mg twice daily.

No pre-existing AF stratum: history of AF an exclusion criterion

Inclusion criteria:

Development of postoperative atrial fibrillation following elective on-pump coronary artery bypass grafting

Exclusion criteria:

History of AF
Prior antiarrhythmic therapy

Patient characteristics:

CABG surgery in all patients
Mean (SD) LVEF: 52.6 (8.6) vs. 53.8 (9.4)%
Diabetes: 40 vs. 38%
Renal insufficiency: 15 vs. 14.28%

Achievement of sinus rhythm

For both groups:

All patients after extubation and until discharge received a standard drug regimen that included acetylsalicylic acid (100 mg daily), atorvastatin (20–40 mg daily), the beta-blocker metoprolol (50–100 mg daily), and the angiotensin-converting enzyme inhibitor perindopril (5–10 mg daily), in addition to each patient’s individual treatment based on his or her personal medical history

Simonpoulos 2018¹⁰⁰

RCT

N=812

Conducted in Greece

K+ blockers

IV amiodarone. Loading dose of 300 mg in 30 min followed by 750 mg in 24 h.

If arrhythmia was sustained after 24 h, further 375 mg given in 12 h. Maximum recording period of 36 h.

After conversion to sinus rhythm amiodarone infusion was discontinued and treatment with amiodarone 200 mg t.i.d was continued until hospital discharge

K+ blockers + ranolazine

IVamiodarone + oral ranolazine. Amiodarone loading dose of 300 mg in 30 min followed by 750 mg in 24 h.

If arrhythmia was sustained after 24 h, further 375 mg given in 12 h. Maximum recording period of 36 h.

500 mg ranolazine was administered once at time of randomisation, followed by 375 mg 6 h later and subsequently 375 mg twice daily.

After conversion to sinus rhythm amiodarone infusion was discontinued and treatment with amiodarone 200 mg t.i.dand375 mg b.i.dranolazine was continued until hospital discharge

No pre-existing AF stratum: persistent or permanent AF in previous 6 months exclusion criterion

Inclusion criteria:

Those that underwent coronary artery bypass grafting
Development of atrial fibrillation 2–3 days following operation

Exclusion criteria:

Previously documented persistent of permanent AF in last 6 months prior to surgery
ReceivingCYP3A inhibitors or inducers
History of hepatic or renal failure

Patient characteristics:

All underwent CABG surgery
Mean (SD) LVEF: 42.65 (8.98) vs. 43.24 (9.7)%
Prior myocardial infarction: 53.8 vs. 59%
Type II diabetes: 54.6 vs. 53.8%
Hypertension: 58.5 vs. 54.3%
Medications:
Beta-blockers: 84.0 vs. 81.8%
Digoxin: 0.2 vs. 0%
ACE inhibitors/ARBs: 65.9 vs. 66.8%
Statins: 58.3 vs. 68.6%
Dihydropyridines: 24.7 vs. 32.9%

Achievement of sinus rhythm

For both groups:

On first postoperative day, all given LMWH and acetylsalicylic acid 100 mg once daily, which was continued during the AF and conversion to sinus rhythm. Where sinus rhythm was not restored with 36 h, anticoagulation was changed to acenocoumarol4 mg and adjusted according to INR

Vilvanathan 2016¹¹⁰

RCT

N=89

Conducted in India

K+ blockers + DC cardioversion

Amiodarone + DC cardioversion. DC cardioversion performed 48 h after balloon mitral valvuloplasty.

Cardioversion:

Prior to DC cardioversion, patients sedated and given analgesia.
Synchronised DC cardioversion was given using biphasic defibrillators using the following protocol: 100J, 200J, 300J and 360 J.
Unsuccessful DC cardioversion was considered to include those who did not revert with 360J.

K+ blockers:

Amiodarone was given as an intravenous bolus of 150 mg followed by a 1 g intravenous infusion for 12 h prior to DC cardioversion.
Following cardioversion, oral amiodarone was started initially 200 mg three times a day for 2 weeks, followed by 200 mg twice daily for 2 weeks and subsequently 200 mg once daily for 12 months

Placebo + DC cardioversion

Cardioversion:

DC cardioversion was performed 48 h after balloon mitral valvuloplasty.
Prior to DC cardioversion, patients were sedated and received analgesia.
Synchronised DC cardioversion was given using biphasic defibrillators using the following protocol: 100J, 200J, 300J and 360 J.
Unsuccessful DC cardioversion was considered to include those who did not revert with 360J.

Placebo:

No preloading with amiodarone prior to DC cardioversion
Following DC cardioversion, patients received placebo for 12 months

Pre-existing AF stratum: all had permanent AF for at least 3 months prior to mitral valvuloplasty

Inclusion criteria:

Aged >18 years
Underwent successful balloon mitral valvuloplasty
ECG evidence of atrial fibrillation for >3 months

Exclusion criteria:

Prior history of cardioversion
Significant mitral, tricuspid or aortic regurgitation
Significant tricuspid or aortic stenosis
Left atrial thrombus (detected by transoesophageal echocardiography)
Left atrial diameter ≥6 cm
Inability to comply with 12 months follow-up period
Contraindications to anticoagulation or amiodarone

Patient characteristics:

Mean (SD) duration of AF: 10.05 (5.718) vs. 10.27 (5.495) months
Hypertension: 4.5 vs. 6.7%
Diabetes: 4.5 vs. 2.2%
Hypertension + diabetes: 2.3 vs. 4.4%
Diabetes + coronary heart disease: 2.3 vs. 0%
NYHA class: I (20.5 vs. 4.4%), II (63.6 vs. 75.6%), III (11.4 vs. 13.3%) and IV (4.5 vs. 6.7%)
Concomitant drugs:
- Beta-blockers: 48.9 vs. 43.2%
- Calcium channel blockers: 24.4 vs. 22.7%
- Digoxin: 64.4 vs. 68.2%

Health-related quality of life
Achievement of sinus rhythm
Adverse events

For both groups:

Patients were anticoagulated with warfarin and INR was required to be between 2 and 3 for at least 1 month prior to DC cardioversion

Wafa 1989¹¹¹

RCT

N=29

Conducted in UK

Na+ blockers

Intravenous flecainide. Bolus of 1 mg/kg body weight over 10 min followed by an infusion of 1.5 mg/kg/h for 1 h and another infusion of 0.25 mg/kg/h for the rest of the 24 h study period

A single dose of verapamil (10 mg intravenously) was given over a 5 min period if after 45 min reversion to sinus rhythm and adequate ventricular rate control (<100 beats/min) had not been achieved.

Digoxin

Intravenous digoxin. Bolus of 0.5 mg over 10 min followed after 6 and 12 h by bolus doses of 0.25 mg over 10 min

A single dose of verapamil (10 mg intravenously) was given over a 5 min period if after 45 min reversion to sinus rhythm and adequate ventricular rate control (<100 beats/min) had not been achieved.

No pre-existing AF: preoperative atrial arrhythmias an exclusion criterion

Inclusion criteria:

18–80 years of age
CABG complicated within 96 h after surgery by atrial tachyarrhythmias (atrial fibrillation, atrial flutter or atrial tachycardia) lasting at least 15 min with a ventricular rate >120 beats/min

Exclusion criteria:

Preoperative atrial arrhythmia
Second- or third-degree atrioventricular block
Presence or history of bifascicular block or bundle branch block with any degree of atrioventricular block
Known sinus node dysfunction in the absence of a pacing wire
Impaired left ventricular dysfunction (as detected clinically and angiographically)
Treatment with other antiarrhythmics (including verapamil) during anaesthesia or since return to intensive care unit
Treatment with digoxin or beta-blockers in the 24h before entry into the study
Serious renal or hepatic disease
Receipt of any investigational drug during the 4-weeks prior to the study
Receipt of any antiarrhythmic agents within 3 elimination half-lives of the date of inclusion in the study

Patient characteristics:

Type of operation: CABG alone (93.3 vs. 92.9%) and CABG + aortic valve replacement (6.7 vs. 7.1%)
Arrhythmia type: atrial fibrillation (100 vs. 85.7%) and atrial flutter (0 vs. 14.3%)
Coronary artery disease: 100% vs. 100%
Aortic valve disease: 6.7 vs. 7.1%

Achievement of sinus rhythm
Adverse events

Table 3Clinical evidence summary: Evidence not suitable for GRADE analysis

Study

Intervention and comparator

Outcome

Intervention results

Intervention group (n)

Comparatorresults

Comparatorgroup (n)

Risk of bias

Gillinov2016³⁵

No pre-existing AF stratum: Mixed rate control vs. K+ blocker with/without rate control agent

Hospital length of stay at 60 days

Median (IQR): 5.1 (3–7.4) days

262

Median (IQR): 5.0 (3.2–7.5) days

261

High

Hwang1984⁵¹

No pre-existing AF stratum: Calcium channel blockers vs. placebo

Achievement of sinus rhythm at end of hospital stay – note results given ‘as received’.

All patients in placebo group switched to verapamil as treatment failed within required time-frame. Study reports total number of all participants in study that were in sinus rhythm after treatment with verapamil

At end of study,3/14 remained in sinus rhythm at the end of their hospital stay. Of these, 2 had been receiving oral doses of digoxin or propanololhydro chloride, but none maintained on oral doses of verapamil

14 (those originally assigned to calcium channel blockers and all those assigned to placebo switched following failure)

Very high

Table 4Clinical evidence summary: Mixed/unclear stratum– calcium channel blockers vs. placebo

Outcomes

No of Participants (studies)

Follow up

Quality of the evidence (GRADE)

Relative effect (95% CI)

Anticipated absolute effects

Risk with placebo

Risk difference with Calcium channel blockers (95% CI)

Adverse events (adverse reaction or unusual haemodynamic response)

(1 study)

hours

⊕⊝⊝⊝

VERY LOW^b^,^c^,^d

due to risk of bias, indirectness, imprecision

RD: 0 (−0.16 to 0.16)

Moderate

0 per 1000

0 fewer per 1000

(from 160 fewer to 160 more)^a

a: Absolute effect calculated manually using risk difference as zero events in both arms of the study.

b: Downgraded by 1 increment if the majority of the evidence was at high risk of bias, and downgraded by 2 increments if the majority of the evidence was at very high risk of bias

c: >10% with atrial flutter rather than atrial fibrillation

d: Imprecision assessed using sample size as zero events in both arms of the study. Sample size <70 so very serious imprecision.

Table 5Clinical evidence summary: No pre-existing AFstratum– DC cardioversion vs. K+ blockers

Outcomes	No of Participants (studies) Follow up	Quality of the evidence (GRADE)	Relative effect (95% CI)	Anticipated absolute effects
Outcomes	No of Participants (studies) Follow up	Quality of the evidence (GRADE)	Relative effect (95% CI)	Risk with K+ blockers	Risk difference with DC cardioversion (95% CI)
Achievement of sinus rhythm (sinus rhythm at 24 h)	18 (1 study) 24 hours	⊕⊝⊝⊝ VERY LOW^a^,^b due to risk of bias, imprecision	RR 0.33 (0.09 to 1.23)	Moderate
Achievement of sinus rhythm (sinus rhythm at 24 h)	18 (1 study) 24 hours	⊕⊝⊝⊝ VERY LOW^a^,^b due to risk of bias, imprecision	RR 0.33 (0.09 to 1.23)	667 per 1000	447 fewer per 1000 (from 607 fewer to 153 more)
Need for rescue DC cardioversion (need for transthoracic cardioversion post-24 h - follow-up unclear)	11 (1 study)	⊕⊝⊝⊝ VERY LOW^a^,^b due to risk of bias, imprecision	PetoOR 0.16 (0 to 8.19)	Moderate
	11 (1 study)	⊕⊝⊝⊝ VERY LOW^a^,^b due to risk of bias, imprecision	PetoOR 0.16 (0 to 8.19)	167 per 1000	167 fewer per 1000 (from 543 fewer to 209 more)^c

a: Downgraded by 1 increment if the majority of the evidence was at high risk of bias, and downgraded by 2 increments if the majority of the evidence was at very high risk of bias

b: Downgraded by 1 increment if the confidence interval crossed one MID or by 2 increments if the confidence interval crossed both MIDs

c: Absolute effect calculated manually using risk difference as zero events in one arm of single study.

Table 6Clinical evidence summary: No pre-existing AFstratum– K+ blockers vs. digoxin

Outcomes	No of Participants (studies) Follow up	Quality of the evidence (GRADE)	Relative effect (95% CI)	Anticipated absolute effects
Outcomes	No of Participants (studies) Follow up	Quality of the evidence (GRADE)	Relative effect (95% CI)	Risk with Digoxin	Risk difference with K+ blockers (95% CI)
Achievement of sinus rhythm (sinus rhythm at 24 h)	30 (1 study) 24 hours	⊕⊝⊝⊝ VERY LOW^a^,^b due to risk of bias, imprecision	RR 1.17 (0.88 to 1.55)	Moderate
Achievement of sinus rhythm (sinus rhythm at 24 h)	30 (1 study) 24 hours	⊕⊝⊝⊝ VERY LOW^a^,^b due to risk of bias, imprecision	RR 1.17 (0.88 to 1.55)	800 per 1000	136 more per 1000 (from 96 fewer to 440 more)
Adverse events (clinically significant hypotension or cardiac conduction abnormalities)	30 (1 study) 24 hours	⊕⊝⊝⊝ VERY LOW^a^,^d due to risk of bias, imprecision	RD: 0 (−0.12 to 0.12)	Moderate
	30 (1 study) 24 hours	⊕⊝⊝⊝ VERY LOW^a^,^d due to risk of bias, imprecision	RD: 0 (−0.12 to 0.12)	0 per 1000	0 fewer per 1000 (from 120 fewer to 120 more)^c
Need for rescue DC cardioversion (direct current reversion post24 h - follow-up unclear)	30 (1 study)	⊕⊝⊝⊝ VERY LOW^a^,^b due to risk of bias, imprecision	PetoOR 0.14 (0 to 6.82)	Moderate
	30 (1 study)	⊕⊝⊝⊝ VERY LOW^a^,^b due to risk of bias, imprecision	PetoOR 0.14 (0 to 6.82)	67 per 1000	67 fewer per 1000 (from 233 fewer to 100 more)^e

a: Downgraded by 1 increment if the majority of the evidence was at high risk of bias, and downgraded by 2 increments if the majority of the evidence was at very high risk of bias

b: Downgraded by 1 increment if the confidence interval crossed one MID or by 2 increments if the confidence interval crossed both MIDs

c: Absolute effect calculated manually using risk difference as zero events in both arms of the study.

d: Imprecision assessed using sample size as zero events in both arms of the study. Sample size <70so very serious imprecision.

e: Absolute effect calculated manually using risk difference as zero events in one arm of the study.

Table 7Clinical evidence summary: No pre-existing AF stratum– K+ blockers vs. K+ blockers + ranolazine

Outcomes

No of Participants (studies)

Follow up

Quality of the evidence (GRADE)

Relative effect (95% CI)

Anticipated absolute effects

Risk with K+ blocker + ranolazine

Risk difference with K+ blockers alone (95% CI)

Achievement of sinus rhythm (sinus rhythm at 36 h/unclear)

853

(2 studies)

⊕⊕⊝⊝

LOW^a

due to risk of bias

RR 1 (0.99 to 1.01)

Moderate

100 per 1000

0 fewer per 1000

(from 10 fewer to 10 more)

a: Downgraded by 1 increment if the majority o fthe evidence was at high risk of bias, and downgraded by 2 increments if the majority of the evidence was at very high risk of bias

Table 8Clinical evidence summary: No pre-existing AF stratum– mixed rate control vs. K+ blockers with/without rate control agent

Outcomes	No of Participants (studies) Follow up	Quality of the evidence (GRADE)	Relative effect (95% CI)	Anticipated absolute effects
Outcomes	No of Participants (studies) Follow up	Quality of the evidence (GRADE)	Relative effect (95% CI)	Risk with K+ blocker with/without rate control agent	Risk difference with mixed rate control (95% CI)
Achievement of sinus rhythm (sinus rhythm at hospital discharge)	518 (1 study)	⊕⊝⊝⊝ VERY LOW^a^,^b due to risk of bias, indirectness	RR 0.96 (0.91 to 1.01)	Moderate
	518 (1 study)	⊕⊝⊝⊝ VERY LOW^a^,^b due to risk of bias, indirectness	RR 0.96 (0.91 to 1.01)	934 per 1000	37 fewer per 1000 (from 84 fewer to 9 more)
Adverse events (serious and nonserious adverse events, other than cerebrovascular/non-cerebral thromboembolism)	523 (1 study) 60 days	⊕⊝⊝⊝ VERY LOW^a^,^b^,^e due to risk of bias, indirectness, imprecision	Rate ratio 0.96 (0.77 to 1.2)	Patients could have more than one event included in rate count. Event rate per 100 patient-months: 31.4.	1.3 fewer per 100 patient-months(from 8.22 fewer to 5.58 more)^c^,^d
Freedom from anticoagulation (no warfarin prescription at hospital discharge)	523 (1 study)	⊕⊝⊝⊝ VERY LOW^a^,^b due to risk of bias, indirectness	RR 1.01 (0.87 to 1.17)	Moderate
	523 (1 study)	⊕⊝⊝⊝ VERY LOW^a^,^b due to risk of bias, indirectness	RR 1.01 (0.87 to 1.17)	567 per 1000	6 more per 1000 (from 74 fewer to 96 more)
Mortality (mortality at 60 days)	523 (1 study) 60 days	⊕⊝⊝⊝ VERY LOW^a^,^b^,^e due to risk of bias, indirectness, imprecision	PetoOR 1.49 (0.26 to 8.66)	Moderate
Mortality (mortality at 60 days)	523 (1 study) 60 days		PetoOR 1.49 (0.26 to 8.66)	8 per 1000	4 more per 1000 (from 6 fewer to 57 more)
Need for rescue DC cardioversion (direct current cardioversion at 60 days)	523 (1 study) 60 days	⊕⊝⊝⊝ VERY LOW^a^,^b^,^e due to risk of bias, indirectness, imprecision	RR 0.66 (0.41 to 1.08)	Moderate
	523 (1 study) 60 days		RR 0.66 (0.41 to 1.08)	138 per 1000	47 fewer per 1000 (from 81 fewer to 11 more)
Rehospitalisation, all-cause (readmission due to any cause at 60 days)	523 (1 study) 60 days	⊕⊝⊝⊝ VERY LOW^a^,^b^,^e due to risk of bias, indirectness, imprecision	Rate ratio 1.0 (0.73to 1.37)	Patients could have more than one event included in rate count. Event rate per 100 patient-months: 18.5.	0 fewer per 100-patient-months(from 5.77fewer to 5.74 more)^c^,^d
Rehospitalisation for AF (readmission due to treatment of AF at 60 days)	523 (1 study) 60 days	⊕⊝⊝⊝ VERY LOW^a^,^b^,^e due to risk of bias, indirectness, imprecision	Rate ratio 0.67 (0.31 to 1.42)	Patients could have more than one event included in rate count. Event rate per 100 patient-months: 3.9	1.3 fewer per 100 patient-months(from 3.71 fewer to 1.12 more)^c^,^d
Stroke or thromboembolic complications (serious and non-serious cerebrovascular, inc, stroke and TIA, and/or non-cerebral thromboembolism at 60 days)	523 (1 study) 60 days	⊕⊝⊝⊝ VERY LOW^a^,^b^,^e due to risk of bias, indirectness, imprecision	Rate ratio 2.33 (0.6to 9.02)	Patients could have more than one event included in rate count. Event rate per 100 patient-months: 0.6	0.8more per 100 patient-months(from 0.44fewer to 2.04 more)^c^,^d

a: Downgraded by 1 increment if the majority of the evidence was at high risk of bias, and downgraded by 2 increments if the majority of the evidence was at very high risk of bias

b: Unclear which rate control agents were included - could include some not listed in our protocol

c: Per100 patient-months.

d: Absolute effect calculated manually using difference in rates per 100 patient months

e: Downgraded by 1 increment if the confidence interval crossed one MID or by 2 increments if the confidence interval crossed both MIDs

Table 9Clinical evidence summary: No pre-existing AF stratum– mixed rhythm control +/- electrical cardioversion vs. mixed rate control

Outcomes	No of Participants (studies) Follow up	Quality of the evidence (GRADE)	Relative effect (95% CI)	Anticipated absolute effects
Outcomes	No of Participants (studies) Follow up	Quality of the evidence (GRADE)	Relative effect (95% CI)	Risk with mixed rate control	Risk difference with mixed rhythm control +/- electrical cardioversion (95% CI)
Achievement of sinus rhythm (sinus rhythm at 8 weeks post-hospital discharge)	50 (1 study) 8 weeks post-hospital discharge	⊕⊝⊝⊝ VERY LOW^a^,^b due to risk of bias, indirectness	RR 1.05 (0.91 to 1.22)	Moderate
	50 (1 study) 8 weeks post-hospital discharge	⊕⊝⊝⊝ VERY LOW^a^,^b due to risk of bias, indirectness	RR 1.05 (0.91 to 1.22)	913 per 1000	46 more per 1000 (from 82 fewer to 201 more)
Mortality (mortality at 8 weeks post-hospital discharge)	50 (1 study) 8 weeks post-hospital discharge	⊕⊝⊝⊝ VERY LOW^a^,^b^,^d due to risk of bias, indirectness, imprecision	PetoOR 6.62 (0.4 to 109.94)	Moderate
Mortality (mortality at 8 weeks post-hospital discharge)	50 (1 study) 8 weeks post-hospital discharge		PetoOR 6.62 (0.4 to 109.94)	0 per 1000	74 more per 1000 (from 46 fewer to 194 more)^c
Hospital length of stay (from surgery to discharge)	50 (1 study)	⊕⊝⊝⊝ VERY LOW^a^,^b due to risk of bias, indirectness	NA	The mean hospital length of stay (from surgery to discharge) in the control groups was 9.7 days	The mean hospital length of stay (from surgery to discharge) in the intervention groups was 2.3 lower (2.72 to 1.88 lower) Note: MID was deemed to be 0.5days (based on 0.5 x median sd[1.0] in mixed rate control group)

a: Downgraded by 1 increment if the majority of the evidence was at high risk of bias, and downgraded by 2 increments if the majority of the evidence was at very high risk of bias

b: Serious indirectness as some in the mixed rate control arm could have received intravenous diltiazem - not available in UK in this form. Proportion unclear.

c: Absolute effect calculated manually using risk difference as zero events in control group

d: Downgraded by 1 increment if the confidence interval crossed one MID or by 2 increments if the confidence interval crossed both MIDs

Table 10Clinical evidence summary: No pre-existing AF stratum– Na+ blockers vs. digoxin

Outcomes	No of Participants (studies) Follow up	Quality of the evidence (GRADE)	Relative effect (95% CI)	Anticipated absolute effects
Outcomes	No of Participants (studies) Follow up	Quality of the evidence (GRADE)	Relative effect (95% CI)	Risk with digoxin	Risk difference with Na+ blockers (95% CI)
Achievement of sinus rhythm (sinus rhythm at 24 h)	29 (1 study) 24 hours	⊕⊝⊝⊝ VERY LOW^a^,^b due to risk of bias, imprecision	RR 1.31 (0.91 to 1.87)	Moderate
Achievement of sinus rhythm (sinus rhythm at 24 h)	29 (1 study) 24 hours	⊕⊝⊝⊝ VERY LOW^a^,^b due to risk of bias, imprecision	RR 1.31 (0.91 to 1.87)	714 per 1000	221 more per 1000 (from 64 fewer to 621 more)
Adverse events (adverse reactions at 24 h)	29 (1 study) 24 hours	⊕⊝⊝⊝ VERY LOW^a^,^b due to risk of bias, imprecision	PetoOR 8.02 (0.76 to 84.1)	Moderate
Adverse events (adverse reactions at 24 h)	29 (1 study) 24 hours	⊕⊝⊝⊝ VERY LOW^a^,^b due to risk of bias, imprecision	PetoOR 8.02 (0.76 to 84.1)	0 per 1000	200 more per 1000 (from 22 fewer to 422 more)^c

a: Downgraded by 1 increment if the majority of the evidence was at high risk of bias, and downgraded by 2 increments if the majority of the evidence was at very high risk of bias

b: Downgraded by 1 increment if the confidence interval crossed one MID or by 2 increments if the confidence interval crossed both MIDs

c: Absolute effect calculated manually using risk difference as zero events in control group

Table 11Clinical evidence summary: No pre-existing AF stratum– Na+ blockers vs. K+ blockers

Outcomes	No of Participants (studies) Follow up	Quality of the evidence (GRADE)	Relative effect (95% CI)	Anticipated absolute effects
Outcomes	No of Participants (studies) Follow up	Quality of the evidence (GRADE)	Relative effect (95% CI)	Risk with K+ blockers	Risk difference with Na+ blockers (95% CI)
Achievement of sinus rhythm (without electrical cardioversion at end of study - includes those switching drug)	122 (1 study)	⊕⊕⊝⊝ LOW^a due to risk of bias	RR 1.01 (0.94 to 1.09)	Moderate
	122 (1 study)	⊕⊕⊝⊝ LOW^a due to risk of bias	RR 1.01 (0.94 to 1.09)	955 per 1000	10 more per 1000 (from 57 fewer to 86 more)
Adverse events (significant side effects at end of study)	122 (1 study)	⊕⊝⊝⊝ VERY LOW^a^,^c due to risk of bias, imprecision	RD: 0 (−0.03 to 0.03)	Moderate
Adverse events (significant side effects at end of study)	122 (1 study)	⊕⊝⊝⊝ VERY LOW^a^,^c due to risk of bias, imprecision	RD: 0 (−0.03 to 0.03)	0 per 1000	0 fewer per 1000 (from 30 fewer to 30 more)^b
Need for rescue DC cardioversion (cardioversion at end of study)	122 (1 study)	⊕⊝⊝⊝ VERY LOW^a^,^d due to risk of bias, imprecision	RR 0.81 (0.14 to 4.69)	Moderate
	122 (1 study)	⊕⊝⊝⊝ VERY LOW^a^,^d due to risk of bias, imprecision	RR 0.81 (0.14 to 4.69)	45 per 1000	9 fewer per 1000 (from 39 fewer to 166 more)

a: Downgraded by 1 increment if the majority of the evidence was at high risk of bias, and downgraded by 2 increments if the majority of the evidence was at very high risk of bias

b: Absolute effect calculated manually using risk difference as zero events in both arms of the study.

c: Imprecision assessed using sample size as zero events in both arms of the study. Sample size >70 and <350 so serious imprecision.

d: Downgraded by 1 increment if the confidence interval crossed one MID or by 2 increments if the confidence interval crossed both MIDs

Table 12Clinical evidence summary: No pre-existing AFstratum– Calcium channel blockers vs. placebo

Outcomes

No of Participants (studies)

Follow up

Quality of the evidence (GRADE)

Relative effect (95% CI)

Anticipated absolute effects

Risk with placebo

Risk difference with calcium channel blockers (95% CI)

Adverse events (adverse events requiring premature termination of study, such as hypotension or bradycardia - in-hospital)

(1 study)

⊕⊝⊝⊝

VERY LOW^b^,^c^,^d

due to risk of bias, indirectness, imprecision

RD 0 (−0.24 to 0.24)

Moderate

0 per 1000

0 fewer per 1000

(from 240 fewer to 240 more)^a

a: Absolute effect calculated manually using risk difference as zero events in both arms of the study.

b: Downgraded by 1 increment if the majority of the evidence was at high risk of bias, and downgraded by 2 increments if the majority of the evidence was at very high risk of bias

c: >10% with atrial flutter rather than atrial fibrillation

d: Imprecision assessed using sample size as zero events in both arms of the study. Sample size <70 so very serious imprecision.

Table 13Clinical evidence summary: No pre-existing AFstratum– K+ blockers (vernakalant) vs. placebo

Outcomes	No of Participants (studies) Follow up	Quality of the evidence (GRADE)	Relative effect (95% CI)	Anticipated absolute effects
Outcomes	No of Participants (studies) Follow up	Quality of the evidence (GRADE)	Relative effect (95% CI)	Risk with placebo	Risk difference with No pre-existing AF stratum: K+ blockers (vernakalant) (95% CI)
Mortality	160 (1 study) 30 days	⊕⊕⊝⊝ LOW^b^,^c due to risk of bias, imprecision	RD 0 (−0.03 to 0.03)	Moderate
Mortality	160 (1 study) 30 days	⊕⊕⊝⊝ LOW^b^,^c due to risk of bias, imprecision	RD 0 (−0.03 to 0.03)	0 per 1000	0 fewer per 1000 (from 30 fewer to 30 more)^a
Achievement of sinus rhythm	161 (1 study) 90 min	⊕⊕⊝⊝ LOW^b due to risk of bias	RR 3.03 (1.54 to 5.94)	Moderate
Achievement of sinus rhythm	161 (1 study) 90 min	⊕⊕⊝⊝ LOW^b due to risk of bias	RR 3.03 (1.54 to 5.94)	148 per 1000	300 more per 1000 (from 80 more to 731 more)
Serious adverse events	161 (1 study) 30 days	⊕⊝⊝⊝ VERY LOW^b^,^d due to risk of bias, imprecision	RR 0.84 (0.32 to 2.19)	Moderate
Serious adverse events	161 (1 study) 30 days	⊕⊝⊝⊝ VERY LOW^b^,^d due to risk of bias, imprecision	RR 0.84 (0.32 to 2.19)	111 per 1000	18 fewer per 1000 (from 75 fewer to 132 more)
Treatment-emergent adverse events	161 (1 study) 24 h	⊕⊝⊝⊝ VERY LOW^b^,^d due to risk of bias, imprecision	RR 1.22 (0.77 to 1.93)	Moderate
Treatment-emergent adverse events	161 (1 study) 24 h	⊕⊝⊝⊝ VERY LOW^b^,^d due to risk of bias, imprecision	RR 1.22 (0.77 to 1.93)	315 per 1000	69 more per 1000 (from 72 fewer to 293 more)

a: Absolute risk calculated manually using risk difference as zero events in both arms of the study

b: Downgraded by 1 increment if the majority of the evidence was at high risk of bias, and downgraded by 2 increments if the majority of the evidence was at very high risk of bias

c: Imprecision assessed using sample size as zero events in both arms of the study. Sample size >70 and <350 so serious imprecision

d: Downgraded by 1 increment if the confidence interval crossed one MID or by 2 increments if the confidence interval crossed both MIDs

Table 14Clinical evidence summary: No pre-existing AF stratum– K+ blockers (amiodarone) vs. routine medical treatment alone

Outcomes	No of Participants (studies) Follow up	Quality of the evidence (GRADE)	Relative effect (95% CI)	Anticipated absolute effects
Outcomes	No of Participants (studies) Follow up	Quality of the evidence (GRADE)	Relative effect (95% CI)	Risk with routine medical treatment alone	Risk difference with K+ blockers (amiodarone) (95% CI)
Achievement of sinus rhythm	84 (1 study) unclear	⊕⊝⊝⊝ VERY LOW^a^,^b due to risk of bias, imprecision	RR 1.43 (0.84 to 2.43)	Moderate
Achievement of sinus rhythm	84 (1 study) unclear	⊕⊝⊝⊝ VERY LOW^a^,^b due to risk of bias, imprecision	RR 1.43 (0.84 to 2.43)	333 per 1000	143 more per 1000 (from 53 fewer to 476 more)
Hospital length of stay	84 (1 study)	⊕⊕⊕⊝ MODERATE^a due to risk of bias		The mean hospital length of stay in the control groups was 14.07 days	The mean hospital length of stay in the intervention groups was 3.83 lower (4.32 to 3.34 lower) Note: MID was deemed to be 0.59days (based on 0.5 x median sd[1.17] in routine medical treatment group)
ICU length of stay	84 (1 study)	⊕⊕⊕⊝ MODERATE^a due to risk of bias		The mean ICU length of stay in the control groups was 2.83 days	The mean ICU length of stay in the intervention groups was 1.14 lower (1.54 to 0.74 lower) Note: MID was deemed to be 0.48days (based on 0.5 x median sd[0.95] in routine medical treatment group)

a: Downgraded by 1 increment if the majority of the evidence was at high risk of bias, and downgraded by 2 increments if the majority of the evidence was at very high risk of bias

b: Downgraded by 1 increment if the confidence interval crossed one MID or by 2 increments if the confidence interval crossed both MIDs

Table 15Clinical evidence summary: Pre-existing AF stratum– DC cardioversion vs. K+ blockers + captopril + simvastatin

Outcomes	No of Participants (studies) Follow up	Quality of the evidence (GRADE)	Relative effect (95% CI)	Anticipated absolute effects
Outcomes	No of Participants (studies) Follow up	Quality of the evidence (GRADE)	Relative effect (95% CI)	Risk with K+ blockers+ captopril + simvastatin	Risk difference with DC cardioversion (95% CI)
Achievement of sinus rhythm (sinus rhythm at end of treatment)	115 (1 study)	⊕⊕⊕⊝ MODERATE^a due to risk of bias	RR 3.67 (2.38 to 5.67)	Moderate
	115 (1 study)	⊕⊕⊕⊝ MODERATE^a due to risk of bias	RR 3.67 (2.38 to 5.67)	268 per 1000	716 more per 1000 (from 370 more to 1000 more)
Adverse events (severe complications at follow-up)	115 (1 study) 3–34 months	⊕⊝⊝⊝ VERY LOW^a^,^c due to risk of bias, imprecision	RD 0 (−0.03 to 0.03)	Moderate
Adverse events (severe complications at follow-up)	115 (1 study) 3–34 months	⊕⊝⊝⊝ VERY LOW^a^,^c due to risk of bias, imprecision	RD 0 (−0.03 to 0.03)	0 per 1000	0 fewer per 1000 (from 30 fewer to 30 more)^b
Adverse events (severe cough during treatment)	115 (1 study)	⊕⊝⊝⊝ VERY LOW^a^,^e due to risk of bias, imprecision	PetoOR 0.13 (0.01 to 2.04)	Moderate
Adverse events (severe cough during treatment)	115 (1 study)	⊕⊝⊝⊝ VERY LOW^a^,^e due to risk of bias, imprecision	PetoOR 0.13 (0.01 to 2.04)	36 per 1000	36 fewer per 1000 (from 94 fewer to 22 more)^d
Adverse events (sinus bradycardia with heart rate of 43–52 bpm during treatment)	115 (1 study)	⊕⊝⊝⊝ VERY LOW^a^,^e due to risk of bias, imprecision	PetoOR 7.14 (0.44 to 115.75)	Moderate
	115 (1 study)	⊕⊝⊝⊝ VERY LOW^a^,^e due to risk of bias, imprecision	PetoOR 7.14 (0.44 to 115.75)	0 per 1000	34 more per 1000 (from 22 fewer to 90 more)^f
Mortality (mortality at follow-up)	115 (1 study) 3–34 months	⊕⊕⊝⊝ LOW^a^,^c due to risk of bias, imprecision	RD 0 (−0.03 to 0.03)	Moderate
Mortality (mortality at follow-up)	115 (1 study) 3–34 months	⊕⊕⊝⊝ LOW^a^,^c due to risk of bias, imprecision	RD 0 (−0.03 to 0.03)	0 per 1000	0 fewer per 1000 (from 30 fewer to 30 more)^b
Rehospitalisation for AF (recurrence of AF at follow-up)	73 (1 study) 3–34 months	⊕⊝⊝⊝ VERY LOW^a^,^e due to risk of bias, imprecision	RR 0.52 (0.05 to 5.33)	Moderate
Rehospitalisation for AF (recurrence of AF at follow-up)	73 (1 study) 3–34 months	⊕⊝⊝⊝ VERY LOW^a^,^e due to risk of bias, imprecision	RR 0.52 (0.05 to 5.33)	67 per 1000	32 fewer per 1000 (from 64 fewer to 290 more)

a: Downgraded by 1 increment if the majority of the evidence was at high risk of bias, and downgraded by 2 increments if the majority of the evidence was at very high risk of bias

b: Absolute effect calculated manually using risk difference as zero events in both arms of the study

c: Imprecision assessed using sample size as zero events in both arms. Sample size >70 and <350 so serious imprecision.

d: Absolute effect calculated manually using risk difference as zero events in one arm of single study

e: Downgraded by 1 increment if the confidence interval crossed one MID or by 2 increments if the confidence interval crossed both MIDs

f: Absolute effect calculated manually using risk difference as zero events in the control group

Table 16Clinical evidence summary: Pre-existing AF stratum– Mixed rate control vs. K+ blockers + captopril + simvastatin

Outcomes	No of Participants (studies) Follow up	Quality of the evidence (GRADE)	Relative effect (95% CI)	Anticipated absolute effects
Outcomes	No of Participants (studies) Follow up	Quality of the evidence (GRADE)	Relative effect (95% CI)	Risk with K+ blockers+ captopril + simvastatin	Risk difference with mixed rate control (95% CI)
Achievement of sinus rhythm (sinus rhythm conversion at 12 months)	99 (1 study) 12 months	⊕⊕⊕⊝ MODERATE^a due to risk of bias	RR 0.15 (0.05 to 0.49)	Moderate
	99 (1 study) 12 months	⊕⊕⊕⊝ MODERATE^a due to risk of bias	RR 0.15 (0.05 to 0.49)	388 per 1000	330 fewer per 1000 (from 198 fewer to 369 fewer)
Adverse events (adverse events requiring discontinuation of one or more study drugs at 12 months)	98 (1 study) 12 months	⊕⊕⊝⊝ LOW^a due to risk of bias	PetoOR 0.11 (0.02 to 0.52)	Moderate
	98 (1 study) 12 months	⊕⊕⊝⊝ LOW^a due to risk of bias	PetoOR 0.11 (0.02 to 0.52)	146 per 1000	146 fewer per 1000 (from 250 fewer to 42 more)^b

a: Downgraded by 1 increment if the majority of the evidence was at high risk of bias, and downgraded by 2 increments if the majority of the evidence was at very high risk of bias

b: Absolute effect calculated manually using risk difference as zero events in one arm of the study

Table 17Clinical evidence summary: Pre-existing AF stratum– K+ blockers + DC cardioversion vs. placebo + DC cardioversion

Outcomes

No of Participa nts (studies)

Follow up

Quality of the evidence (GRADE)

Relative effect (95% CI)

Anticipated absolute effects

Risk with placebo + DC cardioversion

Risk difference with K+ blockers + DC cardioversion (95% CI)

Achievement of sinus rhythm (sinus rhythm at 12 months)

(1 study)

12 months

⊕⊕⊝⊝

LOW^a

due to risk of bias

RR 3.23 (1.58 to 6.61)

Moderate

189 per 1000

421 more per 1000

(from 110 more to 1000 more)

Adverse events (dose reduction due to adverse events at 12 months)

(1 study)

12 months

⊕⊕⊝⊝

LOW^a

due to risk of bias

PetoOR 9.25 (2.35 to 36.43)

Moderate

0 per 1000

205 more per 1000

(from 82 more to 328 more)^b

Health-related quality of life (mental component score SF-8 at 12 months)

Scale from: 0 to 100.

(1 study)

12 months

⊕⊝⊝⊝

VERY LOW^a^,^c

due to risk of bias, imprecision

The mean health-related quality of life (mental component score sf-8 at 12 months) in the control groups was

50.15

Baseline value, mean (SD, n): 43.94

(5.276, n=45)

The mean health-related quality of life (mental component score sf-8 at 12 months) in the intervention groups was

3.74 higher

(1.1 to 6.38 higher)

Baseline value, mean (SD, n): 45.08

(4.928, n=44)

Note: MID was deemed to be 2.61(based on 0.5 x median sd[5.216] in placebo + DC cardioversion group)

Health-related quality of life (physical component score SF-8 at 12 months)

Scale from: 0 to 100.

(1 study)

12 months

⊕⊝⊝⊝

VERY LOW^a^,^d

due to risk of bias, imprecision

The mean health-related quality of life (physical component score sf-8 at 12 months) in the control groups was

46.62

Baseline value, mean (SD, n): 46.46

(4.628, n=45)

The mean health-related quality of life (physical component score sf-8 at 12 months) in the intervention groups was

3.17 higher

(0.24 to 6.1 higher)

Baseline value, mean (SD, n): 48.03

(5.005, n=44);

Note: MID was deemed to be 2.96(based on 0.5 x median sd[5.917] in placebo + DC cardioversion group)

a: Downgraded by 1 increment if the majority of the evidence was at high risk of bias, and downgraded by 2 increments if the majority of the evidence was at very high risk of bias

b: Absolute effect calculated manually using risk difference as zero events in control group of study

c: Downgraded by 1 increment as the confidence intervals crossed the upper MID of 2.61

d: Downgraded by 1 increment as the confidence intervals crossed the upper MID of 2.96

Table 18Drug unit costs

Class	Drug (preparation)	Dose range	Cost range per day	Cost range per year
Class IC (Na+ channel blockers)	Disopyramide (capsules)	300mg to 800 mg daily in divided doses	£0.79 to £2.10	£287.22 to £765.92
	Flecainide acetate (tablet)	50mg bd to 300mg daily	£0.16 to £0.26	£59.13 to £93.26
	Propafenone hydrochloride(tablet)	150mg tid to 300mg tid	£0.25 to £0.49	£89.67 to £179.34
Class II (beta-blockers)	Acebutolol (tablet)	0.4g to 1.2 g daily in 2–3 divided doses.	£0.67 to £2	£242.73 to £728.18
	Atenolol (tablet)	50mg to 100mg daily	£0.02 to £0.05	£8.21 to £16.43
	Bisoprolol fumarate (tablet)	5mg to 10mg od	£0.02 to £0.04	£7.69 to £15.38
	Esmolol hydrochloride (IV)	50–200  micrograms/kg/minute	Cost per infusion bag: £89.69^(b)
	Metoprolol tartare (tablet)	50 mg bd to 300mg daily.	£0.06 to £0.10	£20.08 to £34.81
	Nadolol (tablet)	160mg od	£0.43	£156.43
	Propranolol (tablet)	10–40 mg 3–4 times a day	£0.13 to £0.14	£49.01 to £52.40
Class II & III (beta blockers/K+ channel blocker)	Sotalol hydrochloride (tablet)	80 mg to 320 mg daily in 2 divided doses	£0.08 to £0.28	£28.94 to £102.98
Class III (K+ channel blocker)	Amiodarone (tablet)	200mg od	£0.12	£42.50
	Amiodarone (IV infusion)	Maximum 1.2 g per day	£5.87	N/A
	Dronedarone (tablet)	400mg bd	£2.25	£821.25
	Vernakalant (IV infusion)	Maximum 565 mg per day	£327.70 per day	N/A
Class IV (calcium channel blocker)	Diltiazem hydrochloride	120mg to 360mg daily	£0.13 to £0.38	£46.60 to £139.81
	Verapamil hydrochloride (tablet)	40mg to 120 mg tid	£0.06 to £0.14	£20.34 to £52.40
	Verapamil hydrochloride (slow IV injection)	5–10 mg to be given over 2 minutes	£2.16 to £4.33	N/A
Class V (Positive ionotropic drug)	Digoxin (tablet)	125–250 micrograms daily	£0.06 to £0.11	£20.34 to £40.67

(a): Source of cost and dose: BNF⁷, last accessed January 2020. With exception of diltiazem hydrochloride as this is an unlicensed indication. GC expert advice provided for dosage.

(b): Breviblocpre mixed 2.5mg/250ml infusion bags

: Abbreviations: bd: twice daily; IV: intravenous; N/A: not applicable; od: once daily; tid: three times daily.

Table 19Elective inpatient excess bed days cost

Currency Code	Currency Description	Excess Bed Days	National Average Unit Cost
ED26A	Complex Coronary Artery Bypass Graft with CC Score 10+	272	£312
ED26B	Complex Coronary Artery Bypass Graft with CC Score 5–9	157	£312
ED26C	Complex Coronary Artery Bypass Graft with CC Score 0–4	12	£325
ED27A	Major Coronary Artery Bypass Graft with CC Score 10+	4	£352
ED27B	Major Coronary Artery Bypass Graft with CC Score 5–9	110	£377
ED27C	Major Coronary Artery Bypass Graft with CC Score 0–4	17	£311
ED28A	Standard Coronary Artery Bypass Graft with CC Score 10+	381	£304
ED28B	Standard Coronary Artery Bypass Graft with CC Score 5–9	196	£393
ED28C	Standard Coronary Artery Bypass Graft with CC Score 0–4	198	£276
Weighted average			£322

: Source: National reference costs 2017–2018²²

Table 20Non-elective inpatient excess bed days cost

Currency Code	Currency Description	Excess Bed Days	National Average Unit Cost
ED26A	Complex Coronary Artery Bypass Graft with CC Score 10+	422	£346
ED26B	Complex Coronary Artery Bypass Graft with CC Score 5–9	135	£411
ED26C	Complex Coronary Artery Bypass Graft with CC Score 0–4	51	£264
ED27A	Major Coronary Artery Bypass Graft with CC Score 10+	217	£331
ED27B	Major Coronary Artery Bypass Graft with CC Score 5–9	81	£297
ED27C	Major Coronary Artery Bypass Graft with CC Score 0–4	132	£311
ED28A	Standard Coronary Artery Bypass Graft with CC Score 10+	549	£254
ED28B	Standard Coronary Artery Bypass Graft with CC Score 5–9	679	£303
ED28C	Standard Coronary Artery Bypass Graft with CC Score 0–4	362	£385
Weighted average			£318

: Source: National reference costs 2017–2018²²

Treatment strategies for atrial fibrillation after cardiothoracic surgery

Authors

1. Treatment strategies for atrial fibrillation after cardiothoracic surgery

1.1. Review question: What is the most clinical and cost effective treatment strategy (rate or rhythm control,or no treatment) for peoplewith atrial fibrillation after cardiothoracic surgery?

1.2. Introduction

1.3. PICO table

1.4. Methods and process

1.5. Clinical evidence

1.5.1. Included studies

1.5.2. Excluded studies

1.5.3. Summary of clinical studies included in the evidence review

1.5.4. Quality assessment of clinical studies included in the evidence review

1.6. Economic evidence

1.6.1. Included studies

1.6.2. Excluded studies

1.6.3. Unit costs

1.7. The committee’s discussion of the evidence

1.7.1. Interpreting the evidence

1.7.1.1. The outcomes that matter most

1.7.1.2. The quality of the evidence

1.7.1.3. Benefits and harms

No pre-existing AF

Pre-existing AF

1.7.2. Cost effectiveness and resource use

1.7.3. Other factors the committee took into account

References

Appendices

Appendix A. Review protocols

Appendix B. Literature search strategies

Appendix C. Clinical evidence selection

Appendix D. Clinical evidence tables

Appendix E. Forest plots

Appendix F. GRADE tables

Appendix G. Health economic evidence selection

Appendix H. Health economic evidence tables

Appendix I. Excluded studies

Table 1PICO characteristics of review question

Table 2Summary of studies included in the evidence review

Table 3Clinical evidence summary: Evidence not suitable for GRADE analysis

Table 4Clinical evidence summary: Mixed/unclear stratum– calcium channel blockers vs. placebo

Table 5Clinical evidence summary: No pre-existing AFstratum– DC cardioversion vs. K+ blockers

Table 6Clinical evidence summary: No pre-existing AFstratum– K+ blockers vs. digoxin

Table 7Clinical evidence summary: No pre-existing AF stratum– K+ blockers vs. K+ blockers + ranolazine

Table 8Clinical evidence summary: No pre-existing AF stratum– mixed rate control vs. K+ blockers with/without rate control agent

Table 9Clinical evidence summary: No pre-existing AF stratum– mixed rhythm control +/- electrical cardioversion vs. mixed rate control

Table 10Clinical evidence summary: No pre-existing AF stratum– Na+ blockers vs. digoxin

Table 11Clinical evidence summary: No pre-existing AF stratum– Na+ blockers vs. K+ blockers

Table 12Clinical evidence summary: No pre-existing AFstratum– Calcium channel blockers vs. placebo

Table 13Clinical evidence summary: No pre-existing AFstratum– K+ blockers (vernakalant) vs. placebo

Table 14Clinical evidence summary: No pre-existing AF stratum– K+ blockers (amiodarone) vs. routine medical treatment alone

Table 15Clinical evidence summary: Pre-existing AF stratum– DC cardioversion vs. K+ blockers + captopril + simvastatin

Table 16Clinical evidence summary: Pre-existing AF stratum– Mixed rate control vs. K+ blockers + captopril + simvastatin

Table 17Clinical evidence summary: Pre-existing AF stratum– K+ blockers + DC cardioversion vs. placebo + DC cardioversion

Table 18Drug unit costs

Table 19Elective inpatient excess bed days cost

Table 20Non-elective inpatient excess bed days cost