Invited Review
Posterior pericardiotomy in cardiac surgery: systematic review and meta-analysis
Asian Cardiovascular & Thoracic Annals 2015, Vol. 23(3) 354–362 ß The Author(s) 2014 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0218492314541132 aan.sagepub.com
Sadegh Ali-Hassan-Sayegh1, Seyed Jalil Mirhosseini1, Oliver Liakopoulos2, Anton Sabashnikov3, Hamid Reza Dehghan1, Farbod Sedaghat-Hamedani4, Elham Kayvanpour4, Naser Ghaffari5, Vahid Vahabzadeh5, Mahdi Aghabagheri1, Mohammad Reza Mozayan1 and Aron-Frederik Popov3
Abstract This systematic review with meta-analysis sought to determine the impact of posterior pericardiotomy on incidences of atrial fibrillation and supraventricular arrhythmias, pericardial effusion, pleural effusion, tamponade, and the length of hospital stay after cardiac surgery. We searched for randomized controlled trials, using Medline, Embase, Elsevier and Sciences online databases as well as Google Scholar literature. The effect sizes measured were odds ratio for categorical variables and standard mean difference with 95% confidence interval for calculating differences between mean values of hospital stay in intervention and control groups. A value of p < 0.1 for Q test or I2 > 50% indicated significant heterogeneity between the studies. The literature search of all major databases retrieved 20 studies. After screening, 12 suitable trials were identified, which reported outcomes of 2052 patients undergoing cardiac surgery. Posterior pericardiotomy had an odds ratio of 0.33 [95% confidence interval: 0.18–0.61] p < 0.001 for atrial fibrillation; odds ratio 0.32 [0.15–0.67] p ¼ 0.003 for supraventricular arrhythmias; odds ratio 0.09 [0.04–0.19] p ¼ 0.000 for early pericardial effusion and odds ratio 0.04 [0.02–0.08] p < 0.001 for late pericardial effusion; odds ratio 1.64 [1.23–2.20] p ¼ 0.001 for pleural effusion, odds ratio 0.07 [0.02–0.27] p < 0.001 for tamponade, and standard mean difference ¼ 0.01 [0.12 to 0.14] p ¼ 0.8 for hospital stay. Posterior pericardiotomy is a simple intraoperative technique that can improve postoperative clinical outcomes. However, the incidence of pleural effusion associated with posterior pericardiotomy might be higher.
Keywords Atrial fibrillation, cardiac tamponade, length of stay, pericardial effusion, pleural effusion, pericardiectomy
Introduction Pericardial effusion (PE) commonly occurs after cardiovascular surgery. It is often benign and small in amount. However, it can become circumferential and hemodynamically compromising, leading to reduced cardiac output and pericardial tamponade.1,2 Recent studies reported that postoperative PE can provoke the development of atrial fibrillation (AF) which is the most prevalent arrhythmia in the postoperative period.3 AF was reported in 20%–50% of patients following coronary artery bypass grafting (CABG) and in even more after combined CABG and valve surgery.3,4 AF in the early postoperative period may result in
1 Yazd Cardiovascular Research Center, Afshar Hospital, Shahid Sadoughi University of Medical Sciences, Yazd, Iran 2 Department of Thoracic and Cardiovascular Surgery, West German Heart Center, Essen University Hospital, Essen, Germany 3 Department of Cardiothoracic Transplantation and Mechanical Circulatory Support, Royal Brompton & Harefield NHS Foundation Trust, London, UK 4 Department of Medicine III, University of Heidelberg, Heidelberg, Germany 5 Department of Cardiovascular Surgery, Herzchirurgie Klinikum, Karlsruhe, Germany
Corresponding author: Sadegh Ali-Hassan-Sayegh, Cardiovascular Research Center, Afshar Hospital, Jomhouri Blvd, Yazd, Iran. Email: [email protected]
Downloaded from aan.sagepub.com at East Tennessee State University on May 29, 2015
Ali-Hassan-Sayegh et al.
355
short- and long-term morbidity and mortality. Important risk factors predictive of AF include advanced age, male sex, body mass index >30 kgm2, prolonged aortic crossclamp time, preoperative leukocytosis, increased circulating catecholamines, heightened sympathetic and parasympathetic tone, atrial stretch, transcellular fluid and electrolyte shifts, metabolic abnormalities, inflammation, and pericarditis.3,4 Postoperative AF may increase the rates of stroke and thromboembolic events, trigger hemodynamic instability with heart failure, and consequently lead to prolonged intensive care unit and hospital length of stay (LOS), thereby resulting in increased costs.4,5 Prompt pharmacological or electrical cardioversion allows a high rate of successful conversions to sinus rhythm without major clinical complications or significant changes in terms of hospital LOS. However, AF may recur even after hospital discharge, leading to the need for prolonged anticoagulation and further attempts at electrical cardioversion. Furthermore, with severe left ventricular dysfunction or relevant comorbidities, particularly in elderly patients, AF may be a life-threatening complication. The efficacy of pharmacologic interventions in preventing postoperative AF has been extensively investigated.6–8 Administration of antiarrhythmic drugs for prevention or treatment of AF can reduce its incidence and recurrence rate.5 Nevertheless, the use of these drugs is not free of complications and is associated with significant costs.5 Posterior pericardiotomy (PP) is considered a simple surgical method that reduces the risk of postoperative AF because it allows drainage of pericardial fluid into the left pleural space, reducing the incidence of PE that may trigger AF.9 The aim of this study was to assess the impact of PP on early and late clinical outcomes of cardiac surgery. We conducted a metaanalysis of available and relevant randomized controlled trials, to assess the efficacy of PP on incidences of AF and supraventricular arrhythmias, early and late PE, tamponade, pleural effusions, and hospital LOS after cardiac surgery.
Methods A comprehensive literature search was conducted in major electronic databases (Medline/PubMed, Embase, Elsevier, Sciences online database and Google Scholar) from their inception until September 30, 2013, to identify randomized controlled trials that reported the effects of PP on AF, supraventricular arrhythmias, and early and late PE after cardiac surgery. Predefined search terms were: ‘‘posterior pericardiotomy’’, ‘‘pericardiotomy and atrial fibrillation’’, ‘‘arrhythmias’’, ‘‘pericardial effusion and cardiac surgery’’, ‘‘cardiothoracic surgery’’, ‘‘heart surgery’’,
‘‘cardiopulmonary bypass’’, ‘‘CPB’’, ‘‘coronary artery bypass grafting’’, ‘‘CABG’’, ‘‘CAB’’, ‘‘valve surgery’’, and ‘‘valvular surgery’’. There was no language limitation. All references in the trials included were also reviewed to determine additional studies not indexed in common databases. Studies were included in the analysis when they met the following criteria: prospective randomized controlled trials and cohort studies, adult patients (>18 years) undergoing cardiac surgery, comparison of PP with a control group, and reported data on the incidences of postoperative complications such as AF, early and late PE, tamponade, pleural effusion, and hospital LOS. Manuscripts that did not undergo peer review prior to publication were not included. Two researchers (SAHS and SJM) extracted data independently, and discrepancies were resolved via a consensus standardized abstraction checklist used for recording data from each study. Data retrieved from trials included: author’s name, type of surgery (CABG, valve, or combination surgery), study design, sample size, mean age, and sex. The incidences of AF, PE, pleural effusion, tamponade, and hospital LOS were recorded for each group. All primary authors’ definitions for postoperative AF and early and late PE were accepted; they are summarized in Table 1 for each study. The surgical technique of PP has been described elsewhere.17,20 In brief, a longitudinal or circular incision (Table 1) is performed posterior to the phrenic nerve, extending from the left inferior pulmonary vein to the diaphragm prior to discontinuation of cardiopulmonary bypass. Data were analyzed by STATA version 11.0 utilizing METAN and METABIAS modules. The effect sizes measured were odds ratio (OR) with 95% confidence interval (95%CI) for categorical variables. For noncategorical data, the standard mean difference (SMD) was found with a 95% confidence interval for calculating differences in LOS between intervention and control groups. A value of p < 0.1 for Q test or I2 > 50% indicated significant heterogeneity among the studies. Heterogeneity among trials was accounted for by applying a random effect model when indicated. Quality assessment of randomized controlled trials was performed using the Jadad score.7 The Jadad score assesses 3 items including randomization (0–2 points), blinding of study (0–2 points), and withdrawals and dropouts (0–1 points). Higher scores indicate better reporting (high quality ¼ 5; good quality ¼ 3–4; poor quality ¼ 0–2). Results were considered statistically significant at a p value less than 0.05.
Results The literature search retrieved 20 studies from the screened databases. After detailed evaluation, 8 (40%)
Downloaded from aan.sagepub.com at East Tennessee State University on May 29, 2015
356
Asian Cardiovascular & Thoracic Annals 23(3)
Table 1. Definition and monitoring of atrial fibrillation. Definition of posterior pericardiotomy and exclusion criteria
Author
Posterior pericardiotomy: 4-cm longitudinal incision parallel and 1.5 cm posterior to the phrenic nerve, extending from the left inferior pulmonary vein to the diaphragm. Exclusion criteria: hyperthyroidism, chronic obstructive pulmonary disease, renal dysfunction, ventricular aneurysm, severe left ventricular dysfunction, history of AF or previous CABG, and patients with combined valvular disease were not included to obviate disorders that could be associated with an increased incidence of AF. Patients with dense adhesion of the lung and those receiving beta blockers were also excluded. Posterior pericardiotomy: 4-cm longitudinal incision parallel and posterior to the left phrenic nerve, extending from the left inferior pulmonary vein to the diaphragm. Exclusion criteria: hyperthyroidism, chronic obstructive pulmonary disease, renal dysfunction, left ventricular aneurysm, valvular heart disease, and preoperative paroxysmal AF. Patients receiving beta blockers were also excluded. Posterior pericardiotomy: 4–6-cm incision along the posterior length of left phrenic nerve, initiated near the origin of the left inferior pulmonary vein and extended to diaphragm. Exclusion criteria: history of coagulation disorder, renal and hepatic insufficiency, previous open heart surgery, and anticoagulation drugs usage. Posterior pericardiotomy: 4-cm longitudinal incision parallel and posterior to the left phrenic nerve, extending from the left pulmonary vein to the diaphragm. No exclusion criteria. Posterior pericardiotomy: longitudinal incision parallel and 15 mm posterior to the left phrenic nerve, extending from the left inferior pulmonary vein to the diaphragm. Exclusion criteria: coronary artery disease. Posterior pericardiotomy: longitudinal incision parallel and 15 mm posterior to the left phrenic nerve, extending from the left inferior pulmonary vein to the diaphragm. Exclusion criteria: pervious cardiac or thoracic surgeries, left-sided pleural effusion, adhesions, rhythm disturbances such as AF, documented depression and anxiety, hyperthyroidism, left ventricular aneurysm, renal failure, use of beta blockers, and inability to provide informed consent because of neurologic or psychiatric illness. Posterior pericardiotomy: longitudinal incision parallel and posterior to the left phrenic nerve, extending from the left inferior pulmonary vein to the diaphragm. Exclusion criteria: hyperthyroidism, chronic obstructive pulmonary disease, renal dysfunction, left ventricular aneurysm, severe left ventricular dysfunction, combined valvular heart disease, and beta blockers. Posterior pericardiotomy: 4-cm circular incision parallel and posterior to the left phrenic nerve, extending from the left inferior pulmonary vein to the diaphragm. Exclusion criteria: AF and ventricular rhythm problems before surgery, renal failure, left ventricular aneurysm, chronic obstructive pulmonary disease, severe left ventricular dysfunction, hyperthyroidism, valvular heart disease, bleeding disorders, oral anticoagulation and more than 2 drainage tubes installed, and those who required surgery except for isolated CABG. Posterior pericardiotomy: 4-cm posterior pericardial incision below the left inferior pulmonary vein parallel and posterior to the phrenic nerve. No exclusion criteria. Posterior pericardiotomy: Posterior pericardiotomy is a 4 to 6 cm incision along the posterior length of left phrenic nerve and initiated near the origin of left inferior pulmonary vein and extended to diaphragm Exclusion criteria: severe left ventricular function with ejection fraction lower than 25%, or history of AF and patients with combined valvular disease were not included Posterior pericardiotomy: longitudinal incision parallel and posterior to the phrenic nerve, extending from the left inferior pulmonary vein to the diaphragm. No exclusion criteria. Posterior pericardiotomy: longitudinal incision parallel and posterior to the phrenic nerve, extending from the left inferior pulmonary vein to the diaphragm. Exclusion criteria: renal dysfunction, left ventricle aneurysm, severe left ventricle dysfunction, mitral regurgitation, chronic obstructive pulmonary disease, history of AF, endocrine disorders, history of beta-blocker therapy, and dense adhesion in the pericardium or left pleural cavity.
Ekrim10
AF: atrial fibrillation; CABG: coronary artery bypass grafting.
Downloaded from aan.sagepub.com at East Tennessee State University on May 29, 2015
Farsak11
Sadeghpour12
Bakhshandeh13
Erdil14
Cakalagaoglu15
Kuralay16
Kaygin17
Asimakopoulos18
Bolourian19
Mulay20
Arbatli21
Ali-Hassan-Sayegh et al.
357
were excluded due to insufficient reporting of endpoints of interest (n ¼ 5), the manuscript type (case reports, n ¼ 2), or a retrospective study design (n ¼ 1). The final analysis included 12 clinical trials with 2052 patients. In these studies, some information on postoperative clinical outcomes was reported, including: AF (9 studies with 1772 cases), supraventricular arrhythmia (5 studies with 650 cases), early PE (10 studies with 1778 cases), late PE (9 studies with 1678 cases), pleural effusion (7 studies with 1188 cases), tamponade (6 studies with 1075 cases), and hospital LOS (5 studies with 897 cases). Of the 12 studies, 7 were conducted in Turkey, 3 in Iran, and 2 in the United Kingdom. Patient numbers ranged from 80 to 425 and the mean age was 57 years. The type of surgery was CABG in 9 studies (n ¼ 1442, 70.2%), CABG and/or valvular surgery in 2 studies (n ¼ 510, 24.8%), and valve replacement in one study (n ¼ 100, 4.8%). Of the 2052 patients, 1024 were allocated to the treatment (PP) group and 1028 to the control group. The surgical PP techniques (Table 1) were a longitudinal incision (n ¼ 1627, 79.2%) in 11 studies and a circular incision (n ¼ 425, 20.8%) in one. Demographic data and postoperative clinical outcomes are presented in Tables 2 and 3. Incidences of AF were reported in 9 studies (1772 patients); the overall incidence of postoperative AF was 20.3% with a range of 16% to 37.5%. Whereas pooled treatment effect analysis revealed that PP significantly reduced the incidence of postoperative AF with an OR of 0.33 (95%CI: 0.18–0.61, p < 0.001; Figure 1), using a random model, significant heterogeneity was observed between studies (I2 ¼ 78.6% and p < 0.01). Postoperative supraventricular arrhythmias
were mentioned in 5 studies with 650 patients. The average incidence was 24.7% with a range of 28.2% to 36.9%. Pooled analysis indicated that PP may reduce the incidence of supraventricular arrhythmia with an OR of 0.32 (95%CI: 0.15–0.67, p ¼ 0.003; Figure 2) using a random model, however, significant heterogeneity was again observed between studies (I2 ¼ 70.1% and p ¼ 0.01). Incidences of early and late PE were evaluated in 10 studies with 1778 patients and 9 studies with 1678 patients, respectively. The overall incidences of early and late PE were 21.8% and 8.1%, respectively. Pooled analysis using the random effect model revealed that PP significantly reduced the incidence of early PE (OR ¼ 0.09, 95%CI: 0.04–0.19, p ¼ 0.000; Figure 3) with significant heterogeneity (I2 ¼ 73% and p < 0.01). Analysis of the fixed effect model indicated that the incidence of late PE was significantly lower in the PP groups than in the control groups (OR ¼ 0.04, 95%CI: 0.02–0.08, p < 0.001; Figure 4) without heterogeneity among the studies (I2 ¼ 0%; p ¼ 0.9). The overall incidence of early PE was 19.9% with a range of 3% to 32% in 7 studies with 1188 patients. Pooled analysis with the fixed effect model indicated that PP can increase the incidence of PE (OR ¼ 1.64, 95%CI: 1.23–2.20, p ¼ 0.001) without significant heterogeneity (I2 ¼ 0%; p ¼ 0.5). The overall incidence of tamponade was 2.6% in 6 studies with 1075 cases. Pooled treatment effect analysis showed that PP significantly reduced the incidence of tamponade with an OR of 0.07 (95%CI: 0.02–0.27, p < 0.001) using a fixed model, and heterogeneity was not observed among the studies (I2 ¼ 0.0% and p ¼ 0.9). Only 5 randomized controlled trials (897 patients)
Table 2. Demographic characteristics in randomized controlled trials.
Author 10
Ekrim Farsak11 Sadeghpour12 Bakhshandeh13 Erdil14 Cakalagaoglu15 Kuralay16 Kaygin17 Asimakopoulos18 Bolourian19 Mulay20 Arbatli21
Mean age (years)
Male
Hypertension
PP
Control
PP
Control PP
Control PP
Control PP
Control
59.1 8.9 64.2 8.9 60.68 8.49 67.3 8.2 40.9 13.9 63.2 7.67 57 12 58.8 11.3 61 9 59.5 10.7 62 7 62.3 8.1
60.1 3.2 62.8 5.4 60.3 12.6 68.2 9 43.2 15.4 58.82 12.69 61 8 59.0 11.3 61 2 59.6 10.4 62.5 4.9 60.1 9.0
66% 36% 77.5% 38% 46% 80% 77% 50.2% No data 71% 66% 83.3%
58.1% 32% 80% 42% 32% 86% 73% 49.5% No data 71% 70% 74.5%
48% 36% No data 46% No data 62% No data 37.3% 38% 47% No data 59.3%
22% 14.6% 37.5% 47% No data 30% No data 55.6% 10% No data No data 25.9%
87 26 61.4 8.7 No data No data 115.3 44.4 88.04 37.54 No data No data 62 17 94.14 38.18 85.2 3.6 111.56 34.5
52% 41.3% No data 55% No data 68% No data 35.2% 20% 47% No data 61.1%
Diabetes mellitus Cardiopulmonary bypass
20% 17.3% 65% 40% No data 28% No data 54.9% 20% No data No data 25.9%
PP: posterior pericardiotomy.
Downloaded from aan.sagepub.com at East Tennessee State University on May 29, 2015
89 21 57.5 6.1 No data No data 113.9 51.4 91.68 21.69 No data No data 66 17 95.11 38.47 90 33 116.94 32.4
Downloaded from aan.sagepub.com at East Tennessee State University on May 29, 2015
2006/Turkey 2002/Turkey 2011/Iran 2009/Iran
2005/Turkey
2011/Turkey
1999/Turkey 2011/Turkey 1997/UK 2011/Iran 1995/UK 2003/Turkey
Ekrim10 Farsak11 Sadeghpour12 Bakhshandeh13
Erdil14
Cakalagaoglu15
Kuralay16 Kaygin17 Asimakopoulos18 Bolourian19 Mulay20 Arbatli21
100 212 50 87 50 59
50
50
50 75 40 205
Control CABG CABG CABG CABG valve Valve replacement CABG valve CABG CABG CABG CABG CABG CABG
Surgery
Control PP
Control PP
Pleural effusion
9
7
34 62 9 19 14 12
11 40 No data No data 13 10 No data No data 4 18 No data No data
0
4
30
19
21 32 23 67
0
0
0 0 1 0
35 29 1 54 0 59 32 10 46 2 No data No data No data No data No data No data No data No data No data No data 0 3 4 20 No data 7 3 14 28 0
No data No data No data No data No data No data
No data No data No data No data
6 8 2 0
0
0
6
5
data data data data
7 No data No data 7.38 2.23 No data 13.8 8.4
9.58 2.60
7.7 3.7
0 1 No data 0 0 No data No data No data No data No data No data 5.9 4.7
21 0 10 32 0 7 No data No data No No data No data No No data No data No 0 No data No
4
9
3 7 20 37
LOS (days) Control PP
Tamponade
Control PP
Late pericardial effusion
Control PP
Early pericardial effusion Control PP
5 15 8 19 12 9 7 24 10 28 19 13 No data No data No data No data No data No data 53 59 No data No data No data No data
PP
RCT 6 RCT 14 RCT 12 RCT 5 Cohort 4 RCT 7
RCT
RCT
RCT RCT RCT RCT
Study design
Supraventricular arrhythmia
AF: atrial fibrillation; CABG: coronary artery bypass grafting; LOS: length of stay; PP: posterior pericardiotomy; RCT: randomized controlled trial.
100 213 50 87 50 54
50
50
50 75 40 205
Year/Country PP
Author
No. of patients
Postoperative AF
Table 3. Baseline and design characteristics of randomized controlled trials.
2
2 2 2 2
Jadad score
8 No data No data 8.17 2.71 No data 13.3 4.6
2 2 1 2 1 2
9.68 3.36 2
6.9 1.5
No data No data No data 5.5 5.1
Control
Ali-Hassan-Sayegh et al.
359
% Author
OR (95% CI)
Weight
Ekrim
0.26 (0.09, 0.78)
6.23
Farask
0.22 (0.09, 0.55)
10.04
Kuralay
0.12 (0.05, 0.31)
14.75
Kaygin
0.17 (0.09, 0.32)
26.80
Asimakopoulos
1.44 (0.55, 3.80)
3.16
Bolourian
0.22 (0.08, 0.62)
8.27
Mulay
0.22 (0.07, 0.74)
5.95
Bakhshandeh
0.86 (0.56, 1.33)
20.19
Arbatli
0.58 (0.21, 1.61)
4.61
Overall (I-squared = 78.6%, p = 0.000)
0.38 (0.30, 0.49)
100.00
.0492
1
20.3
Figure 1. Forest plot of odds ratios (OR) for treatment with posterior pericardiotomy on the incidence of postoperative atrial fibrillation.
reported on LOS. Mean LOS for all trials was 7.70 5.13 days (7.75 5.3 days for PP and 7.65 4.9 days for the control group). Pooled analysis applying a fixed effects model revealed that PP failed to reduce LOS (SMD 0.01, 95%CI: 0.12 to 0.14; p ¼ 0.8). Also, no heterogeneity was observed in this meta-analysis (I2 ¼ 44.2%; p ¼ 0.1). All included studies had no appropriate methodological quality (mean Jadad score: 2; range 1–2; Table 3).
Discussion A number of studies have been conducted regarding postoperative treatment for supraventricular arrhythmia and AF, and some drugs such as beta blockers and amiodarone, which possess remarkable effectiveness, have been widely prescribed. In this report, we addressed the question of whether a simple intraoperative technique can decrease the need for antiarrhythmic drugs after surgery, reduce arrhythmia and other early and late postoperative complications, and improve clinical outcomes. Our meta-analysis expresses the fact that PP can remarkably diminish AF after heart surgery. The OR of PP for AF prevention was higher than the
OR of beta blockers and amiodarone for AF prevention in previous meta-analyses.22,23 With a simple incision in the pericardium, not only the incidence of postoperative arrhythmia may be decreased but also the need for antiarrhythmic drugs after surgery may be diminished. Many studies stated that PE can be a factor paving the way for AF manifestation. In our study, PP was considered a factor that can remarkably diminish the incidence of early and late PE as well as tamponade, thus contributing to a declining incidence of arrhythmic events. The previous meta-analysis conducted by Biancari and coleagues24 was exclusively focused on 6 papers on the role of PP in CABG. Similar to our findings, these researchers indicated that PP could diminish PE, AF, and postoperative supraventricular arrhythmia. In their meta-analysis, there was considerable heterogeneity in terms of AF, supraventricular arrhythmia, and early PE. However, the heterogeneity-causing factor was not discussed.24 The present meta-analysis also faced heterogeneity in terms of AF, supraventricular arrhythmia, and early PE. However, papers that stated that PP does not have a remarkable diminishing effect on postoperative complications were
Downloaded from aan.sagepub.com at East Tennessee State University on May 29, 2015
360
Asian Cardiovascular & Thoracic Annals 23(3)
% Author
OR (95% CI)
Weight
Ekrim
0.31 (0.12, 0.80)
15.99
Farask
0.26 (0.11, 0.58)
24.32
Kuralay
0.19 (0.09, 0.39)
35.68
Asimakopoulos
1.41 (0.55, 3.59)
7.42
Mulay
0.15 (0.05, 0.50)
16.60
Overall (I-squared = 70.1%, p = 0.010)
0.31 (0.21, 0.45)
100.00
.0478
1
20.9
Figure 2. Forest plot of odds ratios (OR) for treatment with posterior pericardiotomy on the incidence of postoperative supraventricular arrhythmia.
% Author
OR (95% CI)
Weight
Ekrim
0.19 (0.07, 0.52)
5.78
Farask
0.16 (0.07, 0.38)
8.95
Sadeghpour
0.04 (0.01, 0.18)
6.84
Erdil
0.14 (0.04, 0.46)
5.47
Cakalagaoglu
0.01 (0.00, 0.11)
9.45
Kuralay
0.01 (0.00, 0.06)
16.73
Kaygin
0.18 (0.09, 0.36)
13.75
Mulay
0.13 (0.04, 0.42)
5.76
Bakhshandeh
0.00 (0.00, 0.08)
21.07
Arbatli
0.39 (0.18, 0.86)
6.20
Overall (I-squared = 73.0%, p = 0.000)
0.09 (0.07, 0.13)
100.00
.00031
1
3262
Figure 3. Forest plot of odds ratios (OR) for treatment with posterior pericardiotomy on the incidence of early pericardial effusion.
Downloaded from aan.sagepub.com at East Tennessee State University on May 29, 2015
Ali-Hassan-Sayegh et al.
361
% Author
OR (95% CI)
Weight
Ekrim
0.13 (0.01, 2.67)
2.57
Farask
0.06 (0.00, 1.08)
5.52
Sadeghpour
0.03 (0.00, 0.21)
14.46
Erdil
0.04 (0.00, 0.77)
6.98
Cakalagaoglu
0.10 (0.01, 1.95)
3.30
Kuralay
0.02 (0.00, 0.31)
15.86
Kaygin
0.05 (0.01, 0.23)
23.56
Bakhshandeh
0.01 (0.00, 0.18)
27.74
Arbatli
(Excluded)
0.00
Overall (I-squared = 0.0%, p = 0.911)
0.04 (0.02, 0.08)
100.00
.00067
1
1501
Figure 4. Forest plot of odds ratios (OR) for treatment with posterior pericardiotomy on the incidence of late pericardial effusion.
precisely evaluated. Regarding AF, 3 articles were associated with heterogeneity, of which 2 expressed reduced incidence of AF in the PP group, but with no significant difference. In the other article by Asimakopoulos and colleagues,18 a higher prevalence of AF was reported in the PP group, but inclusion and exclusion factors were not clarified, which might be the reason for different outcomes compared to other studies. Another benefit of our study compared to the analysis by Biancari and coleagues24 is that twice the number of randomized controlled trials was included. Moreover, our study examined articles that considered PP in CABG and valve surgery, whereas the other meta-analysis only investigated only PP in CABG. It should also be noted that none of the previous studies were adequate in terms of the quality of methodology, which is vital in terms of achieving appropriate results. This meta-analysis moreover reports that PP may lead to an increased incidence of PE, in line with previous studies. Regarding hospital LOS, no difference was observed between the PP group and the control group, which may be related to the small number of studies reporting hospital LOS. There might be several complications related to PP, such as cardiac herniation or compression of venous grafts by the edges of the PP.
Therefore, this technique should be used carefully in patients in whom posterior wall revascularization was performed, especially by sequential grafting.25 We concluded that PP is a simple intraoperative technique that might increase the incidence of PE but considerably reduce AF and pericardial tamponade. Therefore, this simple intraoperative technique may provide the capacity to decrease antiarrhythmic drug use for AF and supraventricular arrhythmias and avoid emergency repeat thoracotomies, thus ameliorating postoperative outcomes. Funding This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Conflicts of interest statement None declared.
References 1. Ashikhmina EA, Schaff HV, Sinak LJ, et al. Pericardial effusion after cardiac surgery: risk factors, patient profiles, and contemporary management. Ann Thorac Surg 2010; 89: 112–118.
Downloaded from aan.sagepub.com at East Tennessee State University on May 29, 2015
362
Asian Cardiovascular & Thoracic Annals 23(3)
2. Bryan AJ and Angelini GD. Pericardial effusion after open heart surgery. Thorax 1990; 45: 655–656. 3. Omae T and Kanmura Y. Management of postoperative atrial fibrillation. J Anesth 2012; 26: 429–437. 4. Mirhosseini SJ, Ali-Hassan-Sayegh S and Forouzannia SK. What is the exact predictive role of preoperative white blood cell count for new-onset atrial fibrillation following open heart surgery? Saudi J Anaesth 2013; 7: 40–42. 5. Rostagno C. Atrial fibrillation in patients undergoing surgical revascularization: an update on pharmacologic prophylaxis [Review]. Cardiovasc Hematol Agents Med Chem 2012; 10: 325–338. 6. Steinberg BA, Beckley PD, Deering TF, et al. Evaluation and management of the atrial fibrillation patient: a report from the Society of Cardiovascular Patient Care. Crit Pathw Cardiol 2013; 12: 107–115. 7. Dunning J, Nagendran M, Alfieri OR, et al. Guideline for the surgical treatment of atrial fibrillation. Eur J Cardiothorac Surg 2013; 44: 777–791. 8. Bode K, Sommer P, Bollmann A and Hindricks G. Therapeutic management of non-valvular atrial fibrillation. Update 2013. Herz 2013; 38: 743–755. 9. Kaleda VI, McCormack DJ and Shipolini AR. Does posterior pericardiotomy reduce the incidence of atrial fibrillation after coronary artery bypass grafting surgery? Interact Cardiovasc Thorac Surg 2012; 14: 384–389. 10. Ekim H, Kutay V, Hazar A, Akbayrak H, Bas¸el H and Tuncer M. Effects of posterior pericardiotomy on the incidence of pericardial effusion and atrial fibrillation after coronary revascularization. Med Sci Monit 2006; 12: 431–434. 11. Farsak B, Gu¨naydin S, Tokmakog˘lu H, Kandemir O, Yorganciog˘lu C and Zorlutuna Y. Posterior pericardiotomy reduces the incidence of supra-ventricular arrhythmias and pericardial effusion after coronary artery bypass grafting. Eur J Cardiothorac Surg 2002; 22: 278–281. 12. Sadeghpour A, Baharestani B, Ghasemzade Ghotbabady B, Baghaei R and Givhtaje N. Influences of posterior pericardiotomy in early and late postoperative effusion of pericardium. Iranian J Card Surg 2011; 3: 42–43. Available at: http://www.sid.ir/en/VEWSSID/J_pdf/ 1013120110107.pdf. Accessed June 01, 2014. 13. Bakhshandeh AR, Salehi M, Radmehr H, Sattarzadeh R, Nasr AR and Sadeghpour AH. Postoperative pericardial effusion and posterior pericardiotomy: related? Asian Cardiovasc Thorac Ann 2009; 17: 477–479. 14. Erdil N, Nisanoglu V, Kosar F, Erdil FA, Cihan HB and Battaloglu B. Effect of posterior pericardiotomy on early and late pericardial effusion after valve replacement. J Card Surg 2005; 20: 257–260.
15. Cakalagaoglu C, Koksal C, Baysal A, et al. The use of posterior pericardiotomy technique to prevent postoperative pericardial effusion in cardiac surgery. Heart Surg Forum 2012; 15: E84–E89. 16. Kuralay E, Ozal E, Demirkili U and Tatar H. Effect of posterior pericardiotomy on postoperative supraventricular arrhythmias and late pericardial effusion (posterior pericardiotomy). J Thorac Cardiovasc Surg 1999; 118: 492–495. 17. Kaygin MA, Dag O, Gu¨nes¸ M, et al. Posterior pericardiotomy reduces the incidence of atrial fibrillation, pericardial effusion, and length of stay in hospital after coronary artery bypasses surgery. Tohoku J Exp Med 2011; 225: 103–108. 18. Asimakopoulos G, Della Santa R and Taggart DP. Effects of posterior pericardiotomy on the incidence of atrial fibrillation and chest drainage after coronary revascularization: a prospective randomized trial. J Thorac Cardiovasc Surg 1997; 113: 797–799. 19. Bolourian AA, Beheshti Monfared M, Gachkar L, et al. The preventive effects of posterior pericardiotomy on atrial fibrillation after elective coronary artery bypass grafting. Tehran University Med J 2011; 69: 29–35. Available at: http://tumj.tums.ac.ir/browse.php? a_id¼273&sid¼1&slc_lang¼en. Accessed June 01, 2014. 20. Mulay A, Kirk AJ, Angelini GD, Wisheart JD and Hutter JA. Posterior pericardiotomy reduces the incidence of supra-ventricular arrhythmias following coronary artery bypass surgery. Eur J Cardiothorac Surg 1995; 9: 150–152. 21. Arbatli H, Demirsoy E, Aytekin S, et al. The role of posterior pericardiotomy on the incidence of atrial fibrillation after coronary revascularization. J Cardiovasc Surg (Torino) 2003; 44: 713–717. 22. Burgess DC, Kilbron MJ and Keech AC. Interventions for prevention of post-operative atrial fibrillation and its complications after cardiac surgery: a meta-analysis. Eur Heart J 2006; 27: 2846–2857. 23. Marik PE and Fromm R. The efficacy and dosage effect of corticosteroids for the prevention of atrial fibrillation after cardiac surgery: a systematic review. J Crit Care 2009; 24: 458–463. 24. Biancari F and Mahar MA. Meta-analysis of randomized trials on the efficacy of posterior pericardiotomy in preventing atrial fibrillation after coronary artery bypass surgery. J Thorac Cardiovasc Surg 2010; 139: 1158–1161. 25. Yorganciog˘lu C, Farsak B, Tokmakog˘lu H and Gu¨naydin S. An unusual experience with posterior pericardiotomy. Eur J Cardiothorac Surg 2000; 18: 727–728.
Downloaded from aan.sagepub.com at East Tennessee State University on May 29, 2015
Posterior pericardiotomy in cardiac surgery: systematic review and meta-analysis
Asian Cardiovascular & Thoracic Annals 2015, Vol. 23(3) 354–362 ß The Author(s) 2014 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0218492314541132 aan.sagepub.com
Sadegh Ali-Hassan-Sayegh1, Seyed Jalil Mirhosseini1, Oliver Liakopoulos2, Anton Sabashnikov3, Hamid Reza Dehghan1, Farbod Sedaghat-Hamedani4, Elham Kayvanpour4, Naser Ghaffari5, Vahid Vahabzadeh5, Mahdi Aghabagheri1, Mohammad Reza Mozayan1 and Aron-Frederik Popov3
Abstract This systematic review with meta-analysis sought to determine the impact of posterior pericardiotomy on incidences of atrial fibrillation and supraventricular arrhythmias, pericardial effusion, pleural effusion, tamponade, and the length of hospital stay after cardiac surgery. We searched for randomized controlled trials, using Medline, Embase, Elsevier and Sciences online databases as well as Google Scholar literature. The effect sizes measured were odds ratio for categorical variables and standard mean difference with 95% confidence interval for calculating differences between mean values of hospital stay in intervention and control groups. A value of p < 0.1 for Q test or I2 > 50% indicated significant heterogeneity between the studies. The literature search of all major databases retrieved 20 studies. After screening, 12 suitable trials were identified, which reported outcomes of 2052 patients undergoing cardiac surgery. Posterior pericardiotomy had an odds ratio of 0.33 [95% confidence interval: 0.18–0.61] p < 0.001 for atrial fibrillation; odds ratio 0.32 [0.15–0.67] p ¼ 0.003 for supraventricular arrhythmias; odds ratio 0.09 [0.04–0.19] p ¼ 0.000 for early pericardial effusion and odds ratio 0.04 [0.02–0.08] p < 0.001 for late pericardial effusion; odds ratio 1.64 [1.23–2.20] p ¼ 0.001 for pleural effusion, odds ratio 0.07 [0.02–0.27] p < 0.001 for tamponade, and standard mean difference ¼ 0.01 [0.12 to 0.14] p ¼ 0.8 for hospital stay. Posterior pericardiotomy is a simple intraoperative technique that can improve postoperative clinical outcomes. However, the incidence of pleural effusion associated with posterior pericardiotomy might be higher.
Keywords Atrial fibrillation, cardiac tamponade, length of stay, pericardial effusion, pleural effusion, pericardiectomy
Introduction Pericardial effusion (PE) commonly occurs after cardiovascular surgery. It is often benign and small in amount. However, it can become circumferential and hemodynamically compromising, leading to reduced cardiac output and pericardial tamponade.1,2 Recent studies reported that postoperative PE can provoke the development of atrial fibrillation (AF) which is the most prevalent arrhythmia in the postoperative period.3 AF was reported in 20%–50% of patients following coronary artery bypass grafting (CABG) and in even more after combined CABG and valve surgery.3,4 AF in the early postoperative period may result in
1 Yazd Cardiovascular Research Center, Afshar Hospital, Shahid Sadoughi University of Medical Sciences, Yazd, Iran 2 Department of Thoracic and Cardiovascular Surgery, West German Heart Center, Essen University Hospital, Essen, Germany 3 Department of Cardiothoracic Transplantation and Mechanical Circulatory Support, Royal Brompton & Harefield NHS Foundation Trust, London, UK 4 Department of Medicine III, University of Heidelberg, Heidelberg, Germany 5 Department of Cardiovascular Surgery, Herzchirurgie Klinikum, Karlsruhe, Germany
Corresponding author: Sadegh Ali-Hassan-Sayegh, Cardiovascular Research Center, Afshar Hospital, Jomhouri Blvd, Yazd, Iran. Email: [email protected]
Downloaded from aan.sagepub.com at East Tennessee State University on May 29, 2015
Ali-Hassan-Sayegh et al.
355
short- and long-term morbidity and mortality. Important risk factors predictive of AF include advanced age, male sex, body mass index >30 kgm2, prolonged aortic crossclamp time, preoperative leukocytosis, increased circulating catecholamines, heightened sympathetic and parasympathetic tone, atrial stretch, transcellular fluid and electrolyte shifts, metabolic abnormalities, inflammation, and pericarditis.3,4 Postoperative AF may increase the rates of stroke and thromboembolic events, trigger hemodynamic instability with heart failure, and consequently lead to prolonged intensive care unit and hospital length of stay (LOS), thereby resulting in increased costs.4,5 Prompt pharmacological or electrical cardioversion allows a high rate of successful conversions to sinus rhythm without major clinical complications or significant changes in terms of hospital LOS. However, AF may recur even after hospital discharge, leading to the need for prolonged anticoagulation and further attempts at electrical cardioversion. Furthermore, with severe left ventricular dysfunction or relevant comorbidities, particularly in elderly patients, AF may be a life-threatening complication. The efficacy of pharmacologic interventions in preventing postoperative AF has been extensively investigated.6–8 Administration of antiarrhythmic drugs for prevention or treatment of AF can reduce its incidence and recurrence rate.5 Nevertheless, the use of these drugs is not free of complications and is associated with significant costs.5 Posterior pericardiotomy (PP) is considered a simple surgical method that reduces the risk of postoperative AF because it allows drainage of pericardial fluid into the left pleural space, reducing the incidence of PE that may trigger AF.9 The aim of this study was to assess the impact of PP on early and late clinical outcomes of cardiac surgery. We conducted a metaanalysis of available and relevant randomized controlled trials, to assess the efficacy of PP on incidences of AF and supraventricular arrhythmias, early and late PE, tamponade, pleural effusions, and hospital LOS after cardiac surgery.
Methods A comprehensive literature search was conducted in major electronic databases (Medline/PubMed, Embase, Elsevier, Sciences online database and Google Scholar) from their inception until September 30, 2013, to identify randomized controlled trials that reported the effects of PP on AF, supraventricular arrhythmias, and early and late PE after cardiac surgery. Predefined search terms were: ‘‘posterior pericardiotomy’’, ‘‘pericardiotomy and atrial fibrillation’’, ‘‘arrhythmias’’, ‘‘pericardial effusion and cardiac surgery’’, ‘‘cardiothoracic surgery’’, ‘‘heart surgery’’,
‘‘cardiopulmonary bypass’’, ‘‘CPB’’, ‘‘coronary artery bypass grafting’’, ‘‘CABG’’, ‘‘CAB’’, ‘‘valve surgery’’, and ‘‘valvular surgery’’. There was no language limitation. All references in the trials included were also reviewed to determine additional studies not indexed in common databases. Studies were included in the analysis when they met the following criteria: prospective randomized controlled trials and cohort studies, adult patients (>18 years) undergoing cardiac surgery, comparison of PP with a control group, and reported data on the incidences of postoperative complications such as AF, early and late PE, tamponade, pleural effusion, and hospital LOS. Manuscripts that did not undergo peer review prior to publication were not included. Two researchers (SAHS and SJM) extracted data independently, and discrepancies were resolved via a consensus standardized abstraction checklist used for recording data from each study. Data retrieved from trials included: author’s name, type of surgery (CABG, valve, or combination surgery), study design, sample size, mean age, and sex. The incidences of AF, PE, pleural effusion, tamponade, and hospital LOS were recorded for each group. All primary authors’ definitions for postoperative AF and early and late PE were accepted; they are summarized in Table 1 for each study. The surgical technique of PP has been described elsewhere.17,20 In brief, a longitudinal or circular incision (Table 1) is performed posterior to the phrenic nerve, extending from the left inferior pulmonary vein to the diaphragm prior to discontinuation of cardiopulmonary bypass. Data were analyzed by STATA version 11.0 utilizing METAN and METABIAS modules. The effect sizes measured were odds ratio (OR) with 95% confidence interval (95%CI) for categorical variables. For noncategorical data, the standard mean difference (SMD) was found with a 95% confidence interval for calculating differences in LOS between intervention and control groups. A value of p < 0.1 for Q test or I2 > 50% indicated significant heterogeneity among the studies. Heterogeneity among trials was accounted for by applying a random effect model when indicated. Quality assessment of randomized controlled trials was performed using the Jadad score.7 The Jadad score assesses 3 items including randomization (0–2 points), blinding of study (0–2 points), and withdrawals and dropouts (0–1 points). Higher scores indicate better reporting (high quality ¼ 5; good quality ¼ 3–4; poor quality ¼ 0–2). Results were considered statistically significant at a p value less than 0.05.
Results The literature search retrieved 20 studies from the screened databases. After detailed evaluation, 8 (40%)
Downloaded from aan.sagepub.com at East Tennessee State University on May 29, 2015
356
Asian Cardiovascular & Thoracic Annals 23(3)
Table 1. Definition and monitoring of atrial fibrillation. Definition of posterior pericardiotomy and exclusion criteria
Author
Posterior pericardiotomy: 4-cm longitudinal incision parallel and 1.5 cm posterior to the phrenic nerve, extending from the left inferior pulmonary vein to the diaphragm. Exclusion criteria: hyperthyroidism, chronic obstructive pulmonary disease, renal dysfunction, ventricular aneurysm, severe left ventricular dysfunction, history of AF or previous CABG, and patients with combined valvular disease were not included to obviate disorders that could be associated with an increased incidence of AF. Patients with dense adhesion of the lung and those receiving beta blockers were also excluded. Posterior pericardiotomy: 4-cm longitudinal incision parallel and posterior to the left phrenic nerve, extending from the left inferior pulmonary vein to the diaphragm. Exclusion criteria: hyperthyroidism, chronic obstructive pulmonary disease, renal dysfunction, left ventricular aneurysm, valvular heart disease, and preoperative paroxysmal AF. Patients receiving beta blockers were also excluded. Posterior pericardiotomy: 4–6-cm incision along the posterior length of left phrenic nerve, initiated near the origin of the left inferior pulmonary vein and extended to diaphragm. Exclusion criteria: history of coagulation disorder, renal and hepatic insufficiency, previous open heart surgery, and anticoagulation drugs usage. Posterior pericardiotomy: 4-cm longitudinal incision parallel and posterior to the left phrenic nerve, extending from the left pulmonary vein to the diaphragm. No exclusion criteria. Posterior pericardiotomy: longitudinal incision parallel and 15 mm posterior to the left phrenic nerve, extending from the left inferior pulmonary vein to the diaphragm. Exclusion criteria: coronary artery disease. Posterior pericardiotomy: longitudinal incision parallel and 15 mm posterior to the left phrenic nerve, extending from the left inferior pulmonary vein to the diaphragm. Exclusion criteria: pervious cardiac or thoracic surgeries, left-sided pleural effusion, adhesions, rhythm disturbances such as AF, documented depression and anxiety, hyperthyroidism, left ventricular aneurysm, renal failure, use of beta blockers, and inability to provide informed consent because of neurologic or psychiatric illness. Posterior pericardiotomy: longitudinal incision parallel and posterior to the left phrenic nerve, extending from the left inferior pulmonary vein to the diaphragm. Exclusion criteria: hyperthyroidism, chronic obstructive pulmonary disease, renal dysfunction, left ventricular aneurysm, severe left ventricular dysfunction, combined valvular heart disease, and beta blockers. Posterior pericardiotomy: 4-cm circular incision parallel and posterior to the left phrenic nerve, extending from the left inferior pulmonary vein to the diaphragm. Exclusion criteria: AF and ventricular rhythm problems before surgery, renal failure, left ventricular aneurysm, chronic obstructive pulmonary disease, severe left ventricular dysfunction, hyperthyroidism, valvular heart disease, bleeding disorders, oral anticoagulation and more than 2 drainage tubes installed, and those who required surgery except for isolated CABG. Posterior pericardiotomy: 4-cm posterior pericardial incision below the left inferior pulmonary vein parallel and posterior to the phrenic nerve. No exclusion criteria. Posterior pericardiotomy: Posterior pericardiotomy is a 4 to 6 cm incision along the posterior length of left phrenic nerve and initiated near the origin of left inferior pulmonary vein and extended to diaphragm Exclusion criteria: severe left ventricular function with ejection fraction lower than 25%, or history of AF and patients with combined valvular disease were not included Posterior pericardiotomy: longitudinal incision parallel and posterior to the phrenic nerve, extending from the left inferior pulmonary vein to the diaphragm. No exclusion criteria. Posterior pericardiotomy: longitudinal incision parallel and posterior to the phrenic nerve, extending from the left inferior pulmonary vein to the diaphragm. Exclusion criteria: renal dysfunction, left ventricle aneurysm, severe left ventricle dysfunction, mitral regurgitation, chronic obstructive pulmonary disease, history of AF, endocrine disorders, history of beta-blocker therapy, and dense adhesion in the pericardium or left pleural cavity.
Ekrim10
AF: atrial fibrillation; CABG: coronary artery bypass grafting.
Downloaded from aan.sagepub.com at East Tennessee State University on May 29, 2015
Farsak11
Sadeghpour12
Bakhshandeh13
Erdil14
Cakalagaoglu15
Kuralay16
Kaygin17
Asimakopoulos18
Bolourian19
Mulay20
Arbatli21
Ali-Hassan-Sayegh et al.
357
were excluded due to insufficient reporting of endpoints of interest (n ¼ 5), the manuscript type (case reports, n ¼ 2), or a retrospective study design (n ¼ 1). The final analysis included 12 clinical trials with 2052 patients. In these studies, some information on postoperative clinical outcomes was reported, including: AF (9 studies with 1772 cases), supraventricular arrhythmia (5 studies with 650 cases), early PE (10 studies with 1778 cases), late PE (9 studies with 1678 cases), pleural effusion (7 studies with 1188 cases), tamponade (6 studies with 1075 cases), and hospital LOS (5 studies with 897 cases). Of the 12 studies, 7 were conducted in Turkey, 3 in Iran, and 2 in the United Kingdom. Patient numbers ranged from 80 to 425 and the mean age was 57 years. The type of surgery was CABG in 9 studies (n ¼ 1442, 70.2%), CABG and/or valvular surgery in 2 studies (n ¼ 510, 24.8%), and valve replacement in one study (n ¼ 100, 4.8%). Of the 2052 patients, 1024 were allocated to the treatment (PP) group and 1028 to the control group. The surgical PP techniques (Table 1) were a longitudinal incision (n ¼ 1627, 79.2%) in 11 studies and a circular incision (n ¼ 425, 20.8%) in one. Demographic data and postoperative clinical outcomes are presented in Tables 2 and 3. Incidences of AF were reported in 9 studies (1772 patients); the overall incidence of postoperative AF was 20.3% with a range of 16% to 37.5%. Whereas pooled treatment effect analysis revealed that PP significantly reduced the incidence of postoperative AF with an OR of 0.33 (95%CI: 0.18–0.61, p < 0.001; Figure 1), using a random model, significant heterogeneity was observed between studies (I2 ¼ 78.6% and p < 0.01). Postoperative supraventricular arrhythmias
were mentioned in 5 studies with 650 patients. The average incidence was 24.7% with a range of 28.2% to 36.9%. Pooled analysis indicated that PP may reduce the incidence of supraventricular arrhythmia with an OR of 0.32 (95%CI: 0.15–0.67, p ¼ 0.003; Figure 2) using a random model, however, significant heterogeneity was again observed between studies (I2 ¼ 70.1% and p ¼ 0.01). Incidences of early and late PE were evaluated in 10 studies with 1778 patients and 9 studies with 1678 patients, respectively. The overall incidences of early and late PE were 21.8% and 8.1%, respectively. Pooled analysis using the random effect model revealed that PP significantly reduced the incidence of early PE (OR ¼ 0.09, 95%CI: 0.04–0.19, p ¼ 0.000; Figure 3) with significant heterogeneity (I2 ¼ 73% and p < 0.01). Analysis of the fixed effect model indicated that the incidence of late PE was significantly lower in the PP groups than in the control groups (OR ¼ 0.04, 95%CI: 0.02–0.08, p < 0.001; Figure 4) without heterogeneity among the studies (I2 ¼ 0%; p ¼ 0.9). The overall incidence of early PE was 19.9% with a range of 3% to 32% in 7 studies with 1188 patients. Pooled analysis with the fixed effect model indicated that PP can increase the incidence of PE (OR ¼ 1.64, 95%CI: 1.23–2.20, p ¼ 0.001) without significant heterogeneity (I2 ¼ 0%; p ¼ 0.5). The overall incidence of tamponade was 2.6% in 6 studies with 1075 cases. Pooled treatment effect analysis showed that PP significantly reduced the incidence of tamponade with an OR of 0.07 (95%CI: 0.02–0.27, p < 0.001) using a fixed model, and heterogeneity was not observed among the studies (I2 ¼ 0.0% and p ¼ 0.9). Only 5 randomized controlled trials (897 patients)
Table 2. Demographic characteristics in randomized controlled trials.
Author 10
Ekrim Farsak11 Sadeghpour12 Bakhshandeh13 Erdil14 Cakalagaoglu15 Kuralay16 Kaygin17 Asimakopoulos18 Bolourian19 Mulay20 Arbatli21
Mean age (years)
Male
Hypertension
PP
Control
PP
Control PP
Control PP
Control PP
Control
59.1 8.9 64.2 8.9 60.68 8.49 67.3 8.2 40.9 13.9 63.2 7.67 57 12 58.8 11.3 61 9 59.5 10.7 62 7 62.3 8.1
60.1 3.2 62.8 5.4 60.3 12.6 68.2 9 43.2 15.4 58.82 12.69 61 8 59.0 11.3 61 2 59.6 10.4 62.5 4.9 60.1 9.0
66% 36% 77.5% 38% 46% 80% 77% 50.2% No data 71% 66% 83.3%
58.1% 32% 80% 42% 32% 86% 73% 49.5% No data 71% 70% 74.5%
48% 36% No data 46% No data 62% No data 37.3% 38% 47% No data 59.3%
22% 14.6% 37.5% 47% No data 30% No data 55.6% 10% No data No data 25.9%
87 26 61.4 8.7 No data No data 115.3 44.4 88.04 37.54 No data No data 62 17 94.14 38.18 85.2 3.6 111.56 34.5
52% 41.3% No data 55% No data 68% No data 35.2% 20% 47% No data 61.1%
Diabetes mellitus Cardiopulmonary bypass
20% 17.3% 65% 40% No data 28% No data 54.9% 20% No data No data 25.9%
PP: posterior pericardiotomy.
Downloaded from aan.sagepub.com at East Tennessee State University on May 29, 2015
89 21 57.5 6.1 No data No data 113.9 51.4 91.68 21.69 No data No data 66 17 95.11 38.47 90 33 116.94 32.4
Downloaded from aan.sagepub.com at East Tennessee State University on May 29, 2015
2006/Turkey 2002/Turkey 2011/Iran 2009/Iran
2005/Turkey
2011/Turkey
1999/Turkey 2011/Turkey 1997/UK 2011/Iran 1995/UK 2003/Turkey
Ekrim10 Farsak11 Sadeghpour12 Bakhshandeh13
Erdil14
Cakalagaoglu15
Kuralay16 Kaygin17 Asimakopoulos18 Bolourian19 Mulay20 Arbatli21
100 212 50 87 50 59
50
50
50 75 40 205
Control CABG CABG CABG CABG valve Valve replacement CABG valve CABG CABG CABG CABG CABG CABG
Surgery
Control PP
Control PP
Pleural effusion
9
7
34 62 9 19 14 12
11 40 No data No data 13 10 No data No data 4 18 No data No data
0
4
30
19
21 32 23 67
0
0
0 0 1 0
35 29 1 54 0 59 32 10 46 2 No data No data No data No data No data No data No data No data No data No data 0 3 4 20 No data 7 3 14 28 0
No data No data No data No data No data No data
No data No data No data No data
6 8 2 0
0
0
6
5
data data data data
7 No data No data 7.38 2.23 No data 13.8 8.4
9.58 2.60
7.7 3.7
0 1 No data 0 0 No data No data No data No data No data No data 5.9 4.7
21 0 10 32 0 7 No data No data No No data No data No No data No data No 0 No data No
4
9
3 7 20 37
LOS (days) Control PP
Tamponade
Control PP
Late pericardial effusion
Control PP
Early pericardial effusion Control PP
5 15 8 19 12 9 7 24 10 28 19 13 No data No data No data No data No data No data 53 59 No data No data No data No data
PP
RCT 6 RCT 14 RCT 12 RCT 5 Cohort 4 RCT 7
RCT
RCT
RCT RCT RCT RCT
Study design
Supraventricular arrhythmia
AF: atrial fibrillation; CABG: coronary artery bypass grafting; LOS: length of stay; PP: posterior pericardiotomy; RCT: randomized controlled trial.
100 213 50 87 50 54
50
50
50 75 40 205
Year/Country PP
Author
No. of patients
Postoperative AF
Table 3. Baseline and design characteristics of randomized controlled trials.
2
2 2 2 2
Jadad score
8 No data No data 8.17 2.71 No data 13.3 4.6
2 2 1 2 1 2
9.68 3.36 2
6.9 1.5
No data No data No data 5.5 5.1
Control
Ali-Hassan-Sayegh et al.
359
% Author
OR (95% CI)
Weight
Ekrim
0.26 (0.09, 0.78)
6.23
Farask
0.22 (0.09, 0.55)
10.04
Kuralay
0.12 (0.05, 0.31)
14.75
Kaygin
0.17 (0.09, 0.32)
26.80
Asimakopoulos
1.44 (0.55, 3.80)
3.16
Bolourian
0.22 (0.08, 0.62)
8.27
Mulay
0.22 (0.07, 0.74)
5.95
Bakhshandeh
0.86 (0.56, 1.33)
20.19
Arbatli
0.58 (0.21, 1.61)
4.61
Overall (I-squared = 78.6%, p = 0.000)
0.38 (0.30, 0.49)
100.00
.0492
1
20.3
Figure 1. Forest plot of odds ratios (OR) for treatment with posterior pericardiotomy on the incidence of postoperative atrial fibrillation.
reported on LOS. Mean LOS for all trials was 7.70 5.13 days (7.75 5.3 days for PP and 7.65 4.9 days for the control group). Pooled analysis applying a fixed effects model revealed that PP failed to reduce LOS (SMD 0.01, 95%CI: 0.12 to 0.14; p ¼ 0.8). Also, no heterogeneity was observed in this meta-analysis (I2 ¼ 44.2%; p ¼ 0.1). All included studies had no appropriate methodological quality (mean Jadad score: 2; range 1–2; Table 3).
Discussion A number of studies have been conducted regarding postoperative treatment for supraventricular arrhythmia and AF, and some drugs such as beta blockers and amiodarone, which possess remarkable effectiveness, have been widely prescribed. In this report, we addressed the question of whether a simple intraoperative technique can decrease the need for antiarrhythmic drugs after surgery, reduce arrhythmia and other early and late postoperative complications, and improve clinical outcomes. Our meta-analysis expresses the fact that PP can remarkably diminish AF after heart surgery. The OR of PP for AF prevention was higher than the
OR of beta blockers and amiodarone for AF prevention in previous meta-analyses.22,23 With a simple incision in the pericardium, not only the incidence of postoperative arrhythmia may be decreased but also the need for antiarrhythmic drugs after surgery may be diminished. Many studies stated that PE can be a factor paving the way for AF manifestation. In our study, PP was considered a factor that can remarkably diminish the incidence of early and late PE as well as tamponade, thus contributing to a declining incidence of arrhythmic events. The previous meta-analysis conducted by Biancari and coleagues24 was exclusively focused on 6 papers on the role of PP in CABG. Similar to our findings, these researchers indicated that PP could diminish PE, AF, and postoperative supraventricular arrhythmia. In their meta-analysis, there was considerable heterogeneity in terms of AF, supraventricular arrhythmia, and early PE. However, the heterogeneity-causing factor was not discussed.24 The present meta-analysis also faced heterogeneity in terms of AF, supraventricular arrhythmia, and early PE. However, papers that stated that PP does not have a remarkable diminishing effect on postoperative complications were
Downloaded from aan.sagepub.com at East Tennessee State University on May 29, 2015
360
Asian Cardiovascular & Thoracic Annals 23(3)
% Author
OR (95% CI)
Weight
Ekrim
0.31 (0.12, 0.80)
15.99
Farask
0.26 (0.11, 0.58)
24.32
Kuralay
0.19 (0.09, 0.39)
35.68
Asimakopoulos
1.41 (0.55, 3.59)
7.42
Mulay
0.15 (0.05, 0.50)
16.60
Overall (I-squared = 70.1%, p = 0.010)
0.31 (0.21, 0.45)
100.00
.0478
1
20.9
Figure 2. Forest plot of odds ratios (OR) for treatment with posterior pericardiotomy on the incidence of postoperative supraventricular arrhythmia.
% Author
OR (95% CI)
Weight
Ekrim
0.19 (0.07, 0.52)
5.78
Farask
0.16 (0.07, 0.38)
8.95
Sadeghpour
0.04 (0.01, 0.18)
6.84
Erdil
0.14 (0.04, 0.46)
5.47
Cakalagaoglu
0.01 (0.00, 0.11)
9.45
Kuralay
0.01 (0.00, 0.06)
16.73
Kaygin
0.18 (0.09, 0.36)
13.75
Mulay
0.13 (0.04, 0.42)
5.76
Bakhshandeh
0.00 (0.00, 0.08)
21.07
Arbatli
0.39 (0.18, 0.86)
6.20
Overall (I-squared = 73.0%, p = 0.000)
0.09 (0.07, 0.13)
100.00
.00031
1
3262
Figure 3. Forest plot of odds ratios (OR) for treatment with posterior pericardiotomy on the incidence of early pericardial effusion.
Downloaded from aan.sagepub.com at East Tennessee State University on May 29, 2015
Ali-Hassan-Sayegh et al.
361
% Author
OR (95% CI)
Weight
Ekrim
0.13 (0.01, 2.67)
2.57
Farask
0.06 (0.00, 1.08)
5.52
Sadeghpour
0.03 (0.00, 0.21)
14.46
Erdil
0.04 (0.00, 0.77)
6.98
Cakalagaoglu
0.10 (0.01, 1.95)
3.30
Kuralay
0.02 (0.00, 0.31)
15.86
Kaygin
0.05 (0.01, 0.23)
23.56
Bakhshandeh
0.01 (0.00, 0.18)
27.74
Arbatli
(Excluded)
0.00
Overall (I-squared = 0.0%, p = 0.911)
0.04 (0.02, 0.08)
100.00
.00067
1
1501
Figure 4. Forest plot of odds ratios (OR) for treatment with posterior pericardiotomy on the incidence of late pericardial effusion.
precisely evaluated. Regarding AF, 3 articles were associated with heterogeneity, of which 2 expressed reduced incidence of AF in the PP group, but with no significant difference. In the other article by Asimakopoulos and colleagues,18 a higher prevalence of AF was reported in the PP group, but inclusion and exclusion factors were not clarified, which might be the reason for different outcomes compared to other studies. Another benefit of our study compared to the analysis by Biancari and coleagues24 is that twice the number of randomized controlled trials was included. Moreover, our study examined articles that considered PP in CABG and valve surgery, whereas the other meta-analysis only investigated only PP in CABG. It should also be noted that none of the previous studies were adequate in terms of the quality of methodology, which is vital in terms of achieving appropriate results. This meta-analysis moreover reports that PP may lead to an increased incidence of PE, in line with previous studies. Regarding hospital LOS, no difference was observed between the PP group and the control group, which may be related to the small number of studies reporting hospital LOS. There might be several complications related to PP, such as cardiac herniation or compression of venous grafts by the edges of the PP.
Therefore, this technique should be used carefully in patients in whom posterior wall revascularization was performed, especially by sequential grafting.25 We concluded that PP is a simple intraoperative technique that might increase the incidence of PE but considerably reduce AF and pericardial tamponade. Therefore, this simple intraoperative technique may provide the capacity to decrease antiarrhythmic drug use for AF and supraventricular arrhythmias and avoid emergency repeat thoracotomies, thus ameliorating postoperative outcomes. Funding This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Conflicts of interest statement None declared.
References 1. Ashikhmina EA, Schaff HV, Sinak LJ, et al. Pericardial effusion after cardiac surgery: risk factors, patient profiles, and contemporary management. Ann Thorac Surg 2010; 89: 112–118.
Downloaded from aan.sagepub.com at East Tennessee State University on May 29, 2015
362
Asian Cardiovascular & Thoracic Annals 23(3)
2. Bryan AJ and Angelini GD. Pericardial effusion after open heart surgery. Thorax 1990; 45: 655–656. 3. Omae T and Kanmura Y. Management of postoperative atrial fibrillation. J Anesth 2012; 26: 429–437. 4. Mirhosseini SJ, Ali-Hassan-Sayegh S and Forouzannia SK. What is the exact predictive role of preoperative white blood cell count for new-onset atrial fibrillation following open heart surgery? Saudi J Anaesth 2013; 7: 40–42. 5. Rostagno C. Atrial fibrillation in patients undergoing surgical revascularization: an update on pharmacologic prophylaxis [Review]. Cardiovasc Hematol Agents Med Chem 2012; 10: 325–338. 6. Steinberg BA, Beckley PD, Deering TF, et al. Evaluation and management of the atrial fibrillation patient: a report from the Society of Cardiovascular Patient Care. Crit Pathw Cardiol 2013; 12: 107–115. 7. Dunning J, Nagendran M, Alfieri OR, et al. Guideline for the surgical treatment of atrial fibrillation. Eur J Cardiothorac Surg 2013; 44: 777–791. 8. Bode K, Sommer P, Bollmann A and Hindricks G. Therapeutic management of non-valvular atrial fibrillation. Update 2013. Herz 2013; 38: 743–755. 9. Kaleda VI, McCormack DJ and Shipolini AR. Does posterior pericardiotomy reduce the incidence of atrial fibrillation after coronary artery bypass grafting surgery? Interact Cardiovasc Thorac Surg 2012; 14: 384–389. 10. Ekim H, Kutay V, Hazar A, Akbayrak H, Bas¸el H and Tuncer M. Effects of posterior pericardiotomy on the incidence of pericardial effusion and atrial fibrillation after coronary revascularization. Med Sci Monit 2006; 12: 431–434. 11. Farsak B, Gu¨naydin S, Tokmakog˘lu H, Kandemir O, Yorganciog˘lu C and Zorlutuna Y. Posterior pericardiotomy reduces the incidence of supra-ventricular arrhythmias and pericardial effusion after coronary artery bypass grafting. Eur J Cardiothorac Surg 2002; 22: 278–281. 12. Sadeghpour A, Baharestani B, Ghasemzade Ghotbabady B, Baghaei R and Givhtaje N. Influences of posterior pericardiotomy in early and late postoperative effusion of pericardium. Iranian J Card Surg 2011; 3: 42–43. Available at: http://www.sid.ir/en/VEWSSID/J_pdf/ 1013120110107.pdf. Accessed June 01, 2014. 13. Bakhshandeh AR, Salehi M, Radmehr H, Sattarzadeh R, Nasr AR and Sadeghpour AH. Postoperative pericardial effusion and posterior pericardiotomy: related? Asian Cardiovasc Thorac Ann 2009; 17: 477–479. 14. Erdil N, Nisanoglu V, Kosar F, Erdil FA, Cihan HB and Battaloglu B. Effect of posterior pericardiotomy on early and late pericardial effusion after valve replacement. J Card Surg 2005; 20: 257–260.
15. Cakalagaoglu C, Koksal C, Baysal A, et al. The use of posterior pericardiotomy technique to prevent postoperative pericardial effusion in cardiac surgery. Heart Surg Forum 2012; 15: E84–E89. 16. Kuralay E, Ozal E, Demirkili U and Tatar H. Effect of posterior pericardiotomy on postoperative supraventricular arrhythmias and late pericardial effusion (posterior pericardiotomy). J Thorac Cardiovasc Surg 1999; 118: 492–495. 17. Kaygin MA, Dag O, Gu¨nes¸ M, et al. Posterior pericardiotomy reduces the incidence of atrial fibrillation, pericardial effusion, and length of stay in hospital after coronary artery bypasses surgery. Tohoku J Exp Med 2011; 225: 103–108. 18. Asimakopoulos G, Della Santa R and Taggart DP. Effects of posterior pericardiotomy on the incidence of atrial fibrillation and chest drainage after coronary revascularization: a prospective randomized trial. J Thorac Cardiovasc Surg 1997; 113: 797–799. 19. Bolourian AA, Beheshti Monfared M, Gachkar L, et al. The preventive effects of posterior pericardiotomy on atrial fibrillation after elective coronary artery bypass grafting. Tehran University Med J 2011; 69: 29–35. Available at: http://tumj.tums.ac.ir/browse.php? a_id¼273&sid¼1&slc_lang¼en. Accessed June 01, 2014. 20. Mulay A, Kirk AJ, Angelini GD, Wisheart JD and Hutter JA. Posterior pericardiotomy reduces the incidence of supra-ventricular arrhythmias following coronary artery bypass surgery. Eur J Cardiothorac Surg 1995; 9: 150–152. 21. Arbatli H, Demirsoy E, Aytekin S, et al. The role of posterior pericardiotomy on the incidence of atrial fibrillation after coronary revascularization. J Cardiovasc Surg (Torino) 2003; 44: 713–717. 22. Burgess DC, Kilbron MJ and Keech AC. Interventions for prevention of post-operative atrial fibrillation and its complications after cardiac surgery: a meta-analysis. Eur Heart J 2006; 27: 2846–2857. 23. Marik PE and Fromm R. The efficacy and dosage effect of corticosteroids for the prevention of atrial fibrillation after cardiac surgery: a systematic review. J Crit Care 2009; 24: 458–463. 24. Biancari F and Mahar MA. Meta-analysis of randomized trials on the efficacy of posterior pericardiotomy in preventing atrial fibrillation after coronary artery bypass surgery. J Thorac Cardiovasc Surg 2010; 139: 1158–1161. 25. Yorganciog˘lu C, Farsak B, Tokmakog˘lu H and Gu¨naydin S. An unusual experience with posterior pericardiotomy. Eur J Cardiothorac Surg 2000; 18: 727–728.
Downloaded from aan.sagepub.com at East Tennessee State University on May 29, 2015
