ORIGINAL R ESEARCH AR TICLE
The Efficacy of Glucocorticoids for the Prevention of Atrial Fibrillation, or Length of Intensive Care Unite or Hospital Stay After Cardiac Surgery: A Meta-Analysis Chao Liu,1,2 Jinghui Wang,3 Daofeng Yiu1 & Kunshen Liu1 1 The First Hospital of Hebei Medical University, Hebei Medical University, Shijiazhuang, China 2 Department of Pharmacology, Hebei Medical University, Shijiazhuang, China 3 The First Affiliated Hospital of Hebei Northern Institute, Hebei Northern Institute, Zhangjiakou, China
Keywords Cardiac surgery; Glucocorticoid; Inflammation; Perioperative atrial fibrillation. Correspondence K. Liu, M.D., Heart Center, The First Hospital of Hebei Medical University, Hebei Medical University, 89 Donggang Road, Shijiazhuang, Hebei Province 050031, China. Tel.: +86-311-8591-7033; Fax: +86-311-8591-7290; E-mail: [email protected]
doi: 10.1111/1755-5922.12062
SUMMARY Aims: cardiopulmonary bypass and cardiac surgery are associated with a significant systemic inflammatory response that has been suggested playing a causative role in the development of perioperative atrial fibrillation (POAF). The goal of this meta-analysis was to determine the efficacy of glucocorticoid prophylaxis in preventing POAF, or length of intensive care unite (ICU) or hospital stay. Methods: A systematic electronic database literature search of Cochrane controlled trials register (2013, issue 4) and MEDLINE (1966 to April, 2013) was conducted using specific search terms for all relevant articles. Including criteria were: randomized controlled clinical trials, participants were adults (≥18 years of age) undergoing cardiac surgery, evaluated glucocorticoid prophylaxis in cardiac surgery with adequately reported data on incidence of POAF, or length of ICU or hospital stay. Results: Forty-two randomized controlled trials involving 7621 participants were included in the meta-analysis. Overall, glucocorticoids prophylaxis significantly lowered participants’ risk of developing POAF (Relative Risk [RR] 0.77; 95% confidence interval [CI] 0.66–0.90), P < 0.01), reduced length of ICU stay by 0.25 day (95% CI 0.40 to 0.10, P < 0.01). Glucocorticoid prophylaxis in cardiac surgery was not associated with increased all-cause of infection (RR 0.68; 95% CI 0.58–0.78, P < 0.01) and mortality (RR 0.75; 95% CI 0.52–1.08, P = 0.12). Conclusions: Pooled evidence suggests that glucocorticoid prophylaxis may reduce the incidence of POAF. The beneficial effect on POAF is associated with reduced length of ICU and hospital stay, and infection rate.
Introduction Perioperative atrial fibrillation (POAF) after cardiac surgery is one of most common complications after cardiac surgery and is associated with significant morbidities [1]. It may lead to hemodynamic instability, prolonged length of intensive care unit (ICU) and hospital stay [2]. Although the use of b-blockers and amiodarone could reduce the risk of POAF, the incidence of POAF remains high. Additional therapeutic options are thus needed. Cardiopulmonary bypass and ischemia/reperfusion injury in cardiac surgery have been shown to provoke a systemic inflammatory response. This systemic inflammatory response may lead to the development of postoperative complications including myocardial injury, respiratory failure, and multiple organ failure. It is well recognized that cardiopulmonary bypass and ischemia/reperfusion injury in cardiac surgery are associated with release of proinflammatory cytokines such as tumor necrosis factor (TNF) alpha, C-reactive protein (CRP), and interleukins (ILs) and repressing these proinflammatory cytokines by glucocorticoids are associated with less occurrence of postoperative complications after cardiac surgery [3–6].
ª 2014 John Wiley & Sons Ltd
Glucocorticoid prophylaxis in cardiac surgery has been studied extensively in past three decades. There are two meta-analyses that assessed the efficacy of glucocorticoid prophylaxis on POAF in cardiac surgery [7,8]. But, both of them included less than 10 randomized clinical trials (RCTs), and sample sizes in most of the included RCTs were small. Many large-scale RCTs have been published ever since and presented controversial results [9–16]. The role of glucocorticoid prophylaxis in preventing POAF remains controversial. Thus, we preformed the present meta-analysis to determine the efficacy of glucocorticoid prophylaxis use on prevention of POAF in patients undergoing cardiac surgery.
Methods Search Strategy We conducted a systematic electronic database literature search of Cochrane controlled trials register (2013, issue 4) and MEDLINE (1966 to April, 2013) for all relevant articles. We used the following Medical Subject Heading (MeSH) terms: steroid, corticosteroid,
Cardiovascular Therapeutics 32 (2014) 89–96
89
Glucocorticoids and Perioperative AF
glucocorticoid and individual drug names (e.g., dexamethasone, prednisolone, prednisone, methylprednisolone, hydrocortisone) with cardiac surgery, cardiopulmonary bypass, or heart surgery. The search was limited to human studies and clinical trials. A manual search of references from reports of clinical trial or review articles was performed to identify relevant trials. Three authors (LC, WJ and YD) independently reviewed all relevant articles. We resolved disagreement by discussion. There were no language restrictions for inclusion in this meta-analysis.
C. Liu et al.
For trials having more than two treatments comparing different doses of glucocorticoid, we broke up the control group into several parts, and the total numbers added up to the original size of the group. If combined treatment data available, we grouped all the treatment groups together and compared them collectively with the control group. Data were analyzed by Review Manager (version 5.01 for Mac; The Cochrane Collaboration, Oxford, UK, 2013). A value of P < 0.05 was regarded as significant in this meta-analysis. Metaregression was performed using metafor package for R.
Selection Criteria The inclusion criteria of trials in the meta-analysis were: randomized controlled clinical trials, participants were adults (≥18 years of age) undergoing cardiac surgery, evaluated glucocorticoids in cardiac surgery with adequately reported data on incidence of POAF, or length of ICU or hospital stay. Studies that used unequal concurrent medical therapies excluded. Two independent reviewers (LC and LK) examined all identified trials to confirm that they fulfilled the inclusion criteria.
Assessment of Risk of Bias in Included Studies Methodological quality of the included trials was assessed against the following criteria: randomization, random allocation, blinding, withdrawals, and analysis according to intention-to-treat. The grading of allocation concealment was based on the Cochrane approach, i.e., adequate, uncertain, or clearly inadequate.
Outcomes The primary outcome was the proportion of patients with newonset POAF in the postoperative period. The secondary outcomes were length of ICU and hospital stay.
Results Characteristics of the Included Studies A total of 42 randomized controlled trails involving 7621 participants were identified and included in the meta-analysis (Figure 1) [3–6,9–46]. All studies were published in English. The dosage of glucocorticoids and the proportion of patients receiving b-blockers to prevent POAF were highly variable between the trials (Table S1). Only 5 out of 42 trials used POAF as their primary outcome. Twelve trials were with high quality design, i.e., trials with adequate concealment and used double or triple blinding.
Effects of Glucocorticoid Prophylaxis on POAF There was a significant heterogeneity between the trials included in the meta-analysis. Therefore, the trials were pooled together using random effects model. Glucocorticoid prophylaxis was associated with a significant reduction in the risk of POAF (RR 0.77; 95% CI 0.66–0.90; P < 0.01; Figure 2). There was no obvious variation in the RR of POAF in the pooled studies. RR of POAF between the low-dose and moderate-dose of glucocorticoid prophylaxis were 0.72 (95% CI 0.60–0.87; P < 0.01) and 0.76 (95% CI 0.60–0.96; P = 0.02), respectively. The effect of high dose of
Statistical Analysis The incidence of POAF was treated as dichotomous data and reported as relative risk (RR) with 95% confidence interval (CI) using a random-effects model. Length of ICU and hospital stay was treated as continuous data and reported as weighted mean difference (WMD) using a random-effects model. We defined total glucocorticoid use less than 1000 mg hydrocortisone or equivalent as low dose, the total glucocorticoid use larger than 10,000 mg hydrocortisone or equivalent as high dose and the total glucocorticoid use between 1000 mg and 10,000 hydrocortisone or equivalent as medium dose. The presence of heterogeneity between trials was assessed by the v2 statistic, and the extent heterogeneity was assessed using I2 statistics. An I2 > 30% was defined as significant heterogeneity in present meta-analysis. Publication bias was assessed by funnel plot with the risk of POAF for the studies whose primary endpoint was POAF. Sensitivity analysis was conducted to assess how the results were affected by restricting the analysis to the trails that were with high quality (i.e., double blinded and with adequate allocation concealment), using b-blockers (≥ 50% participants were on b-blockers), using POAF as primary clinical outcome, or participants underwent coronary artery bypass graft (CABG) surgery.
90
Cardiovascular Therapeutics 32 (2014) 89–96
Figure 1 Flow chart showing study inclusion and exclusion in this metaanalysis.
ª 2014 John Wiley & Sons Ltd
Glucocorticoids and Perioperative AF
C. Liu et al.
Figure 2 The effect of glucocorticoid prophylaxis on risk of atrial fibrillation in adult cardiac surgery.
glucocorticoid prophylaxis on POAF was not conclusive due to limited sample size (RR 0.90; 95% CI 0.57–1.42). Meta-regression showed there was no correlation between dosages of glucocorticoid and the effect of POAF prevention in the pooled studies (Figure 3).
Effects of Glucocorticoid Prophylaxis on Length of ICU and Hospital Stay Only 33 out of 42 published their data on length of ICU stay, and 24 trials reported their length of ICU stay as mean and standard
ª 2014 John Wiley & Sons Ltd
deviation or standard error. Finally, data from 24 trials (n = 1302) were pooled together. The pooled data showed that glucocorticoid prophylaxis reduced length of ICU stay by 0.25 day (95% CI, 0.40, 0.10, P < 0.01; Figure 4). Thirty out of 42 trials reported their data on LOS of hospital stay, and 21 trials involving 1304 participants were pooled together to assess the effects of glucocorticoid prophylaxis on length of hospital stay. It showed that there was a trend that glucocorticoid prophylaxis might reduce length of hospital stay by 0.40 day (95% CI, 1.02, 0.22, P = 0.21; Figure 5). Nine out of 33 trials reported their data on length of ICU stay as median and range or interquartile range (IQR) [4,9–
Cardiovascular Therapeutics 32 (2014) 89–96
91
Glucocorticoids and Perioperative AF
C. Liu et al.
12,16,39,40,43]. Among these trials, 6 trials favor glucocorticoids [4,9,11,16,39,40]; whereas 3 trials reported there was no difference between two groups [10,12,43]. Nine out of thirty trials report their data on length of hospital stay as median and range or IQR. Two trails favor glucocorticoids [4,9]; whereas seven trials reported there was no difference between two groups [10– 12,16,39,40,43]. However, the largest trial involving 4494 participants reported that the mean time to discharge from the ICU was 34.2 h in the dexamethasone group versus 43.6 h in the placebo group (P < 0.001) and mean time of length of hospital stay in the dexamethasone group was 11.3 days versus 11.7 days in the placebo group (P = 0.009), favoring dexamethasone.
Effects of Glucocorticoid Prophylaxis on Adverse Event
Figure 3 Meta-regression using doses of glucocorticoid as a predictor of risk of atrial fibrillation in adult cardiac surgery (slope of the regression line and 95% CI).
Glucocorticoid prophylaxis was not associated with increase of all cause perioperative infection (RR 0.68, 95% CI, 0.58–0.78, n = 6512, P < 0.001). There was a trend that glucocorticoid prophylaxis might reduce mortality related to cardiac surgery (RR, 0.75, 95% CI, 0.52–1.08, n = 6902, P = 0.12).
Figure 4 The effect of glucocorticoid prophylaxis on length of ICU stay in adult cardiac surgery.
92
Cardiovascular Therapeutics 32 (2014) 89–96
ª 2014 John Wiley & Sons Ltd
Glucocorticoids and Perioperative AF
C. Liu et al.
Figure 5 The effect of glucocorticoid prophylaxis on length of hospital stay in adult cardiac surgery.
Sensitivity Analyses and Publication Bias After the analysis was restricted to trials with high quality design, using b-blockers, using POAF as primary clinical outcome, or participants underwent CABG, the effect of glucocorticoid prophylaxis on POAF remained unchanged (Table 1). There were a total of 68 RCTs involving glucocorticoid prophylaxis in adult cardiac surgery with high heterogeneity in their prespecified primary outcomes. Only 26 RCTs reported their data on the occurrence of POAF. Even in those RCTs that reported detailed data on POAF, only 5 RCTs used POAF as the primary outcome. The funnel plot suggested there was no publication bias in the trials used POAF for their primary outcome (Figure 6).
Discussion This meta-analysis showed that glucocorticoid prophylaxis significantly reduced the incidence of atrial fibrillation and length of
ª 2014 John Wiley & Sons Ltd
ICU and hospital stay. This benefit was not significantly different between low, medium and high doses of glucocorticoid. This significant clinical benefit obtained by glucocorticoid prophylaxis further supports the role inflammation in cardiac surgery and potential therapeutic role of anti-inflammatory agents in this setting. The systemic inflammatory response observed in patients undergoing cardiopulmonary bypass and cardiac surgery increases plasma levels of several acute phase proteins such as CRP and TNF are usually short-lived, but can be severe and prolonged in some cases [6,18,20,24,25]. The sustained increase in cytokine levels is associated with adverse outcomes such as cardiovascular events, and stroke and thromboembolism [47]. Glucocorticoids, as anti-inflammatory drugs, are widely used for the treatment of numerous inflammatory diseases. Twenty-two out of 42 RCTs in the present meta-analysis reported the effect of glucocorticoids on inflammatory response [3,5,6,11,14,16,18,20,24–27,32,35, 36,40–46]. All of them unanimously reported that glucocorticoid prophylaxis reduced inflammatory response with a reduction of
Cardiovascular Therapeutics 32 (2014) 89–96
93
Glucocorticoids and Perioperative AF
C. Liu et al.
Table 1 Sensitivity analysis with restriction
Subgroups Primary outcome was POAF b-blockers (≧50%) High quality design Patients with valvular diseases Patients with CABG
Study (n)
Patients (n)
RR with 95% CI
P-value
5
702
0.65 (0.53–0.80)
0.0001
8 13 3
5416 5811 161
0.77 (0.62–0.96) 0.71 (0.58–0.88) 0.39 (0.23–0.66)
0.02 0.002 0.0005
15
1581
0.75 (0.61–0.92)
0.006
Figure 6 The funnel plot for trials using risk of POAF as the primary end point.
CRP, TNF, IL-1, 6, 8 and complement components, which might contribute to the reduction of POAF. Theoretically, anti-inflammatory effects of glucocorticoids are dose-dependent. However, our results did not show that glucocorticoid prophylaxis with high dose was superior over low or medium dose of glucocorticoid prophylaxis in preventing POAF. It has been suggested that high doses of glucocorticoids may alter the phospholipid cell membrane, causing potassium flux across the cell membrane [48]. In humans, intravenous methylprednisolone alters the stimulation threshold of myocardial cells and the urinary excretion of both potassium and sodium. These changes might alter electrolyte shifts across myocardial cell membranes, resulting in cardiac arrhythmias. It is consistent with pervious reports that intravenous use of large doses of glucocorticoids was associated with cardiac arrhythmia [49,50]. In viewing of the fact that most of data are from trials using medium dose of glucocorticoids, more reliance should be on the medium dose of glucocorticoids when used in patients undergoing cardiac surgery. As immunosuppressive drugs, glucocorticoids may increase the risk of infection after surgery. However, our results showed that the infection rate in glucocorticoid group was dramatically lower than that in control group. Short course of use of glucocorticoid prophylaxis (< 72 h) in cardiac surgery did not increase the risk of perioperative infection, but reduced infection rate. Possible explanation is that glucocorticoid prophylaxis could significantly
94
Cardiovascular Therapeutics 32 (2014) 89–96
reduce respiratory failure in patients with cardiac surgery [9]. The beneficial effects obtained by glucocorticoid prophylaxis in post-cardiac arrhythmia, length of ICU and hospital stay, and perioperative infection may be translated into the positive effect on mortality. The majority of the studies (61.8%) did not use POAF as an outcome. Even for the trials that reported occurrence of POAF, only five trials used POAF as their primary outcome. It, inevitably, posed high risk for POAF being missed due to their study design. Therefore, we only funnel plotted the risk of POAF for the trials that used POAF as primary outcome to assess publication bias. The present meta-analysis has several limitations. First, publication bias may be a big concern because many of the studies that did not have POAF as an endpoint were excluded from the analysis. Moreover, the variable duration of follow-up and incomplete monitoring for POAF in some of the included studies may affect the accuracy of the incidence of POAF reported in the present meta-analysis. However, the efficacy of glucocorticoid prophylaxis on POAF remains unchanged when we restricted meta-analysis in the trials that used POAF as their primary outcome. Furthermore, small size trials with negative results other than positive results are difficult to be published, leading to overrepresentation of a drug’s benefit in systematic reviews. However, there are no significant differences in the direction and magnitude of glucocorticoids’ benefit after we restricted the analysis to trials with larger sample size (n ≥ 50) studies. Third, medications such as betablockers and amiodarone proven to be effective at preventing POAF may have been used in the trials. But, their use was seldom reported in the included studies. Nine included trials provided detailed data about beta-blocker use (Table S1). Only one trial clearly reported they did not use amiodarone [43], the data about amiodarone in other 41 trials were not available. Whether adding glucocorticoid on the top of beta-blocker or amiodarone would further reduce POAF incidence remains inconclusive. However, our limited data showed glucocorticoids might further reduce risk of POAF when we pooled the data from the trials that the percentage of beta-blocker use was larger than 50%. Finally, a wide range of doses, and different preparations of glucocorticoid used in the different studies make it harder to identify the ideal dosage and administration time of glucocorticoid in patients undergoing cardiac surgery.
Conclusions The results of present met-analysis suggests that both low-dose of and medium-dose of glucocorticoid could effectively reduce the risk of POAF without increasing the advent events in patients undergoing cardiac surgery. The beneficial effect on POAF is associated with a reduced length of ICU and hospital stay, and infection rate.
Funding Sources China Scholarship Council grant (201208130092).
Conflict of Interest The authors have no conflicts of interest to disclose.
ª 2014 John Wiley & Sons Ltd
Glucocorticoids and Perioperative AF
C. Liu et al.
References
with severe lv dysfunction undergoing elective off-pump coronary artery bypass surgery. Acta
1. Mathew JP, Fontes ML, Tudor IC, et al., Investigators of the Ischemia R, Education F,
Medica Iranica 2011;49:288–292. 14. Kilger E, Heyn J, Beiras-Fernandez A, Luchting
cardiopulmonary bypass. Crit Care Med 2001;29:2137–2142. 26. El Azab SR, Rosseel PM, de Lange JJ, Groeneveld AB, van Strik R, van Wijk EM,
Multicenter Study of Perioperative Ischemia
B, Weis F. Stress doses of hydrocortisone reduce
Scheffer GJ. Dexamethasone decreases the pro-
Research G. A multicenter risk index for atrial
systemic inflammatory response in patients
to anti-inflammatory cytokine ratio during
fibrillation after cardiac surgery. JAMA
undergoing cardiac surgery without
2004;291:1720–1729.
cardiopulmonary bypass. Minerva Anestesiol
2. McKeown P, Epstein AE, American College of Chest P. Future directions: American college of
2011;77:268–274. 15. Yasser Mohamed A, Elmistekawy E, El-Serogy
chest physicians guidelines for the prevention
H. Effects of dexamethasone on pulmonary and
and management of postoperative atrial
renal functions in patients undergoing cabg
fibrillation after cardiac surgery. Chest
with cardiopulmonary bypass. Semin Cardiothorac
2005;128:61S–64S. 3. Fillinger MP, Rassias AJ, Guyre PM, et al.
Vasc Anesth 2009;13:231–237. 16. Weis F, Beiras-Fernandez A, Schelling G, et al.
Glucocorticoid effects on the inflammatory and
Stress doses of hydrocortisone in high-risk
clinical responses to cardiac surgery. J
patients undergoing cardiac surgery: Effects on
Cardiothorac Vasc Anesth 2002;16:163–169.
interleukin-6 to interleukin-10 ratio and
4. Kilger E, Weis F, Briegel J, et al. Stress doses of hydrocortisone reduce severe systemic
cardiac surgery. Crit Care Med 2003;31:1068–1074.
reperfusion phenomena in cardiopulmonary
dexamethasone on the incidence of shivering and recovery in patients undergoing valve replacement surgery. Eg J Anaesth 2003;19:361– 370. 29. Halvorsen P, Raeder J, White PF, Almdahl SM,
revascularization procedures. Anesth Analg 2003;96:1578–1583. 30. Volk T, Schmutzler M, Engelhardt L, et al.
bypass. J Thorac Cardiovasc Surg 1991;102:515–
Effects of different steroid treatment on
525.
reperfusion-associated production of reactive
reaction occurring during cardiopulmonary
18. Engelman RM, Rousou JA, Flack JE III, Deaton
bypass: Effects on tnf-alpha, il-6, il-8, il-10.
DW, Kalfin R, Das DK. Influence of steroids on
Perfusion 2004;19:185–191.
complement and cytokine generation after
prednisolone on serum cytokine levels following
205–214. 28. Abd El-Hakeem EE, Zareh ZE. Effects of
dexamethasone on side effects after coronary
et al. Inhibition by dexamethasone of the
OA, Cengiz M. Effect of low-dose methyl
replacement surgery. Eg J Anaesth 2003;19:
Nordstrand K, Saatvedt K, Veel T. The effect of
early outcome in a risk group of patients after
6. Yilmaz M, Ener S, Akalin H, Sagdic K, Serdar
balance in patients undergoing valve
early outcome. Crit Care Med 2009;37:1685– 17. Jansen NJ, van Oeveren W, van den Broek L,
H. Methylprednisolone prevents inflammatory
Influence of dexamethasone on cytokine
1690.
inflammatory response syndrome and improve
5. Celik JB, Gormus N, Okesli S, Gormus ZI, Solak
cardiac surgery. Br J Anaesth 2002;88:496–501. 27. Abd El-Hakeem EE, El-Minshawy AM.
cardiopulmonary bypass. Ann Thorac Surg 1995;60:801–804. 19. Chaney MA, Nikolov MP, Blakeman B, Bakhos
oxygen species and arrhythmias during coronary surgery. Acta Anaesthesiol Scand 2003;47:667–674. 31. Loef BG, Henning RH, Epema AH, Rietman GW, van Oeveren W, Navis GJ, Ebels T. Effect of dexamethasone on perioperative renal
extracorporeal circulation. Perfusion 1999;
M, Slogoff S. Pulmonary effects of
function impairment during cardiac surgery
14:201–206.
methylprednisolone in patients undergoing
with cardiopulmonary bypass. Br J Anaesth
7. Baker WL, White CM, Kluger J, Denowitz A, Konecny CP, Coleman CI. Effect of perioperative corticosteroid use on the incidence of postcardiothoracic surgery atrial fibrillation and length of stay. Heart Rhythm 2007;4:461–468. 8. Marik PE, Fromm R. The efficacy and dosage
coronary artery bypass grafting and early tracheal extubation. Anesth Analg 1998;87:27– 33. 20. Tassani P, Richter JA, Barankay A, et al. Does high-dose methylprednisolone in aprotinin-treated patients attenuate the systemic
2004;93:793–798. 32. McBride WT, Allen S, Gormley SM, et al. Methylprednisolone favourably alters plasma and urinary cytokine homeostasis and subclinical renal injury at cardiac surgery. Cytokine 2004;27:81–89. 33. Oliver WC Jr, Nuttall GA, Orszulak TA, Bamlet
effect of corticosteroids for the prevention of
inflammatory response during coronary artery
atrial fibrillation after cardiac surgery: A
bypass grafting procedures? J Cardiothorac Vasc
WR, Abel MD, Ereth MH, Schaff HV.
systematic review. J Crit Care 2009;24:
Anesth 1999;13:165–172.
Hemofiltration but not steroids results in earlier
458–463. 9. Dieleman JM, Nierich AP, Rosseel PM, et al.
21. Wan S, LeClerc JL, Huynh CH, Schmartz D, DeSmet JM, Yim AP, Vincent JL. Does steroid
Intraoperative high-dose dexamethasone for
pretreatment increase endotoxin release during
cardiac surgery: A randomized controlled trial.
clinical cardiopulmonary bypass? J Thorac
JAMA 2012;308:1761–1767. 10. Abbaszadeh M, Khan ZH, Mehrani F,
Cardiovasc Surg 1999;117:1004–1008. 22. Yared JP, Starr NJ, Torres FK, et al. Effects of
Jahanmehr H. Perioperative intravenous
single dose, postinduction dexamethasone on
corticosteroids reduce incidence of atrial
recovery after cardiac surgery. Ann Thorac Surg
fibrillation following cardiac surgery: A
2000;69:1420–1424.
randomized study. Rev Bras Cir Cardiovasc
tracheal extubation following cardiopulmonary bypass: A prospective, randomized double-blind trial. Anesthesiology 2004;101:327–339. 34. Bingol H, Cingoz F, Balkan A, Kilic S, Bolcal C, Demirkilic U, Tatar H. The effect of oral prednisolone with chronic obstructive pulmonary disease undergoing coronary artery bypass surgery. J Card Surg 2005;20:252–256. 35. Prasongsukarn K, Abel JG, Jamieson WR,
23. Chaney MA, Durazo-Arvizu RA, Nikolov MP,
Cheung A, Russell JA, Walley KR, Lichtenstein
Blakeman BP, Bakhos M. Methylprednisolone
SV. The effects of steroids on the occurrence of
does not benefit patients undergoing coronary
postoperative atrial fibrillation after coronary
et al. Steroids and statins: An old and a new
artery bypass grafting and early tracheal
artery bypass grafting surgery: A prospective
anti-inflammatory strategy compared. Perfusion
extubation. J Thorac Cardiovasc Surg
randomized trial. J Thorac Cardiovasc Surg
2011;26:31–37.
2001;121:561–569.
2012;27:18–23. 11. Vukovic PM, Maravic-Stojkovic VR, Peric MS,
12. Murphy GS, Sherwani SS, Szokol JW, et al.
24. Schurr UP, Zund G, Hoerstrup SP, Grunenfelder
2005;130:93–98. 36. Rubens FD, Nathan H, Labow R, et al. Effects of
Small-dose dexamethasone improves quality of
J, Maly FE, Vogt PR, Turina MI. Preoperative
methylprednisolone and a biocompatible
recovery scores after elective cardiac surgery: A
administration of steroids: Influence on
copolymer circuit on blood activation during
randomized, double-blind, placebo-controlled
adhesion molecules and cytokines after
cardiopulmonary bypass. Ann Thorac Surg
study. J Cardiothorac Vasc Anesth 2011;25:950–
cardiopulmonary bypass. Ann Thorac Surg
2005;79:655–665.
960. 13. Mirhosseini SJ, Forouzannia SK, Sayegh AH,
2001;72:1316–1320. 25. Volk T, Schmutzler M, Engelhardt L, Docke
37. Enc Y, Karaca P, Ayoglu U, Camur G, Kurc E, Cicek S. The acute cardioprotective effect of
Sanatkar M. Effect of prophylactic low dose of
WD, Volk HD, Konertz W, Kox WJ. Influence
glucocorticoid in myocardial
methylprednisolone on postoperative new atrial
of aminosteroid and glucocorticoid treatment on
ischemia-reperfusion injury occurring during
fibrillation and early complications in patients
inflammation and immune function during
ª 2014 John Wiley & Sons Ltd
Cardiovascular Therapeutics 32 (2014) 89–96
95
Glucocorticoids and Perioperative AF
C. Liu et al.
cardiopulmonary bypass. Heart Vessels
methylprednisolone in patients undergoing
2006;21:152–156.
cardiac surgery. Ann Thorac Surg 2007;84:110–
38. Sano T, Morita S, Masuda M, Yasui H. Minor infection encouraged by steroid administration
118; discussion 118-119 43. Yared JP, Bakri MH, Erzurum SC, et al. Effect
during cardiac surgery. Asian Cardiovasc Thorac
of dexamethasone on atrial fibrillation after
Ann 2006;14:505–510.
cardiac surgery: Prospective, randomized,
39. Weis F, Kilger E, Roozendaal B, et al. Stress doses of hydrocortisone reduce chronic stress symptoms and improve health-related quality of
double-blind, placebo-controlled trial. J Cardiothorac Vasc Anesth 2007;21:68–75. 44. Kilickan L, Yumuk Z, Bayindir O. The effect of
derivative to decrease possible neuronal damage. J Card Surg 2009;24:397–403. 47. Mendall MA, Patel P, Asante M, et al. Relation of serum cytokine concentrations to cardiovascular risk factors and coronary heart disease. Heart 1997;78:273–277. 48. Barnes PJ. How corticosteroids control inflammation: Quintiles prize lecture 2005. Br J Pharmacol 2006;148:245–254.
life in high-risk patients after cardiac surgery: A
combined preinduction thoracic epidural
randomized study. J Thorac Cardiovasc Surg
anaesthesia and glucocorticoid administration
supraventricular tachycardia after intravenous
2006;131:277–282.
on perioperative interleukin-10 levels and
pulse methylprednisolone therapy, with a brief
hyperglycemia. A randomized controlled trial. J
review of the literature. Rheumatol Int
40. Whitlock RP, Young E, Noora J, Farrokhyar F, Blackall M, Teoh KH. Pulse low dose steroids attenuate post-cardiopulmonary bypass sirs; sirs i. J Surg Res 2006;132:188–194.
Cardiovasc Surg 2008;49:87–93. 45. Sobieski MA II, Graham JD, Pappas PS, Tatooles
49. Kumari R, Uppal SS. First report of
2005;26:70–73. 50. Moretti R, Torre P, Antonello RM, Zorzon M,
AJ, Slaughter MS. Reducing the effects of the
Cazzato G. Recurrent atrial fibrillation
systemic inflammatory response to
associated with pulse administration of high
Corticosteroids for the prevention of atrial
cardiopulmonary bypass: Can single dose
doses of methylprednysolone: A possible
fibrillation after cardiac surgery: A randomized
steroids blunt systemic inflammatory response
prophylactic treatment. Eur J Neurol 2000;7:130.
controlled trial. JAMA 2007;297:1562–1567.
syndrome? ASAIO J 2008;54:203–206.
41. Halonen J, Halonen P, Jarvinen O, et al.
42. Liakopoulos OJ, Schmitto JD, Kazmaier S,
46. Demir T, Demir H, Tansel T, et al. Influence of
Brauer A, Quintel M, Schoendube FA, Dorge H.
methylprednisolone on levels of neuron-specific
Cardiopulmonary and systemic effects of
enolase in cardiac surgery: A corticosteroid
Supporting Information Additional Supporting Information may be found in the online version of this article: Table S1 Clinical characteristics of included studies.
96
Cardiovascular Therapeutics 32 (2014) 89–96
ª 2014 John Wiley & Sons Ltd
The Efficacy of Glucocorticoids for the Prevention of Atrial Fibrillation, or Length of Intensive Care Unite or Hospital Stay After Cardiac Surgery: A Meta-Analysis Chao Liu,1,2 Jinghui Wang,3 Daofeng Yiu1 & Kunshen Liu1 1 The First Hospital of Hebei Medical University, Hebei Medical University, Shijiazhuang, China 2 Department of Pharmacology, Hebei Medical University, Shijiazhuang, China 3 The First Affiliated Hospital of Hebei Northern Institute, Hebei Northern Institute, Zhangjiakou, China
Keywords Cardiac surgery; Glucocorticoid; Inflammation; Perioperative atrial fibrillation. Correspondence K. Liu, M.D., Heart Center, The First Hospital of Hebei Medical University, Hebei Medical University, 89 Donggang Road, Shijiazhuang, Hebei Province 050031, China. Tel.: +86-311-8591-7033; Fax: +86-311-8591-7290; E-mail: [email protected]
doi: 10.1111/1755-5922.12062
SUMMARY Aims: cardiopulmonary bypass and cardiac surgery are associated with a significant systemic inflammatory response that has been suggested playing a causative role in the development of perioperative atrial fibrillation (POAF). The goal of this meta-analysis was to determine the efficacy of glucocorticoid prophylaxis in preventing POAF, or length of intensive care unite (ICU) or hospital stay. Methods: A systematic electronic database literature search of Cochrane controlled trials register (2013, issue 4) and MEDLINE (1966 to April, 2013) was conducted using specific search terms for all relevant articles. Including criteria were: randomized controlled clinical trials, participants were adults (≥18 years of age) undergoing cardiac surgery, evaluated glucocorticoid prophylaxis in cardiac surgery with adequately reported data on incidence of POAF, or length of ICU or hospital stay. Results: Forty-two randomized controlled trials involving 7621 participants were included in the meta-analysis. Overall, glucocorticoids prophylaxis significantly lowered participants’ risk of developing POAF (Relative Risk [RR] 0.77; 95% confidence interval [CI] 0.66–0.90), P < 0.01), reduced length of ICU stay by 0.25 day (95% CI 0.40 to 0.10, P < 0.01). Glucocorticoid prophylaxis in cardiac surgery was not associated with increased all-cause of infection (RR 0.68; 95% CI 0.58–0.78, P < 0.01) and mortality (RR 0.75; 95% CI 0.52–1.08, P = 0.12). Conclusions: Pooled evidence suggests that glucocorticoid prophylaxis may reduce the incidence of POAF. The beneficial effect on POAF is associated with reduced length of ICU and hospital stay, and infection rate.
Introduction Perioperative atrial fibrillation (POAF) after cardiac surgery is one of most common complications after cardiac surgery and is associated with significant morbidities [1]. It may lead to hemodynamic instability, prolonged length of intensive care unit (ICU) and hospital stay [2]. Although the use of b-blockers and amiodarone could reduce the risk of POAF, the incidence of POAF remains high. Additional therapeutic options are thus needed. Cardiopulmonary bypass and ischemia/reperfusion injury in cardiac surgery have been shown to provoke a systemic inflammatory response. This systemic inflammatory response may lead to the development of postoperative complications including myocardial injury, respiratory failure, and multiple organ failure. It is well recognized that cardiopulmonary bypass and ischemia/reperfusion injury in cardiac surgery are associated with release of proinflammatory cytokines such as tumor necrosis factor (TNF) alpha, C-reactive protein (CRP), and interleukins (ILs) and repressing these proinflammatory cytokines by glucocorticoids are associated with less occurrence of postoperative complications after cardiac surgery [3–6].
ª 2014 John Wiley & Sons Ltd
Glucocorticoid prophylaxis in cardiac surgery has been studied extensively in past three decades. There are two meta-analyses that assessed the efficacy of glucocorticoid prophylaxis on POAF in cardiac surgery [7,8]. But, both of them included less than 10 randomized clinical trials (RCTs), and sample sizes in most of the included RCTs were small. Many large-scale RCTs have been published ever since and presented controversial results [9–16]. The role of glucocorticoid prophylaxis in preventing POAF remains controversial. Thus, we preformed the present meta-analysis to determine the efficacy of glucocorticoid prophylaxis use on prevention of POAF in patients undergoing cardiac surgery.
Methods Search Strategy We conducted a systematic electronic database literature search of Cochrane controlled trials register (2013, issue 4) and MEDLINE (1966 to April, 2013) for all relevant articles. We used the following Medical Subject Heading (MeSH) terms: steroid, corticosteroid,
Cardiovascular Therapeutics 32 (2014) 89–96
89
Glucocorticoids and Perioperative AF
glucocorticoid and individual drug names (e.g., dexamethasone, prednisolone, prednisone, methylprednisolone, hydrocortisone) with cardiac surgery, cardiopulmonary bypass, or heart surgery. The search was limited to human studies and clinical trials. A manual search of references from reports of clinical trial or review articles was performed to identify relevant trials. Three authors (LC, WJ and YD) independently reviewed all relevant articles. We resolved disagreement by discussion. There were no language restrictions for inclusion in this meta-analysis.
C. Liu et al.
For trials having more than two treatments comparing different doses of glucocorticoid, we broke up the control group into several parts, and the total numbers added up to the original size of the group. If combined treatment data available, we grouped all the treatment groups together and compared them collectively with the control group. Data were analyzed by Review Manager (version 5.01 for Mac; The Cochrane Collaboration, Oxford, UK, 2013). A value of P < 0.05 was regarded as significant in this meta-analysis. Metaregression was performed using metafor package for R.
Selection Criteria The inclusion criteria of trials in the meta-analysis were: randomized controlled clinical trials, participants were adults (≥18 years of age) undergoing cardiac surgery, evaluated glucocorticoids in cardiac surgery with adequately reported data on incidence of POAF, or length of ICU or hospital stay. Studies that used unequal concurrent medical therapies excluded. Two independent reviewers (LC and LK) examined all identified trials to confirm that they fulfilled the inclusion criteria.
Assessment of Risk of Bias in Included Studies Methodological quality of the included trials was assessed against the following criteria: randomization, random allocation, blinding, withdrawals, and analysis according to intention-to-treat. The grading of allocation concealment was based on the Cochrane approach, i.e., adequate, uncertain, or clearly inadequate.
Outcomes The primary outcome was the proportion of patients with newonset POAF in the postoperative period. The secondary outcomes were length of ICU and hospital stay.
Results Characteristics of the Included Studies A total of 42 randomized controlled trails involving 7621 participants were identified and included in the meta-analysis (Figure 1) [3–6,9–46]. All studies were published in English. The dosage of glucocorticoids and the proportion of patients receiving b-blockers to prevent POAF were highly variable between the trials (Table S1). Only 5 out of 42 trials used POAF as their primary outcome. Twelve trials were with high quality design, i.e., trials with adequate concealment and used double or triple blinding.
Effects of Glucocorticoid Prophylaxis on POAF There was a significant heterogeneity between the trials included in the meta-analysis. Therefore, the trials were pooled together using random effects model. Glucocorticoid prophylaxis was associated with a significant reduction in the risk of POAF (RR 0.77; 95% CI 0.66–0.90; P < 0.01; Figure 2). There was no obvious variation in the RR of POAF in the pooled studies. RR of POAF between the low-dose and moderate-dose of glucocorticoid prophylaxis were 0.72 (95% CI 0.60–0.87; P < 0.01) and 0.76 (95% CI 0.60–0.96; P = 0.02), respectively. The effect of high dose of
Statistical Analysis The incidence of POAF was treated as dichotomous data and reported as relative risk (RR) with 95% confidence interval (CI) using a random-effects model. Length of ICU and hospital stay was treated as continuous data and reported as weighted mean difference (WMD) using a random-effects model. We defined total glucocorticoid use less than 1000 mg hydrocortisone or equivalent as low dose, the total glucocorticoid use larger than 10,000 mg hydrocortisone or equivalent as high dose and the total glucocorticoid use between 1000 mg and 10,000 hydrocortisone or equivalent as medium dose. The presence of heterogeneity between trials was assessed by the v2 statistic, and the extent heterogeneity was assessed using I2 statistics. An I2 > 30% was defined as significant heterogeneity in present meta-analysis. Publication bias was assessed by funnel plot with the risk of POAF for the studies whose primary endpoint was POAF. Sensitivity analysis was conducted to assess how the results were affected by restricting the analysis to the trails that were with high quality (i.e., double blinded and with adequate allocation concealment), using b-blockers (≥ 50% participants were on b-blockers), using POAF as primary clinical outcome, or participants underwent coronary artery bypass graft (CABG) surgery.
90
Cardiovascular Therapeutics 32 (2014) 89–96
Figure 1 Flow chart showing study inclusion and exclusion in this metaanalysis.
ª 2014 John Wiley & Sons Ltd
Glucocorticoids and Perioperative AF
C. Liu et al.
Figure 2 The effect of glucocorticoid prophylaxis on risk of atrial fibrillation in adult cardiac surgery.
glucocorticoid prophylaxis on POAF was not conclusive due to limited sample size (RR 0.90; 95% CI 0.57–1.42). Meta-regression showed there was no correlation between dosages of glucocorticoid and the effect of POAF prevention in the pooled studies (Figure 3).
Effects of Glucocorticoid Prophylaxis on Length of ICU and Hospital Stay Only 33 out of 42 published their data on length of ICU stay, and 24 trials reported their length of ICU stay as mean and standard
ª 2014 John Wiley & Sons Ltd
deviation or standard error. Finally, data from 24 trials (n = 1302) were pooled together. The pooled data showed that glucocorticoid prophylaxis reduced length of ICU stay by 0.25 day (95% CI, 0.40, 0.10, P < 0.01; Figure 4). Thirty out of 42 trials reported their data on LOS of hospital stay, and 21 trials involving 1304 participants were pooled together to assess the effects of glucocorticoid prophylaxis on length of hospital stay. It showed that there was a trend that glucocorticoid prophylaxis might reduce length of hospital stay by 0.40 day (95% CI, 1.02, 0.22, P = 0.21; Figure 5). Nine out of 33 trials reported their data on length of ICU stay as median and range or interquartile range (IQR) [4,9–
Cardiovascular Therapeutics 32 (2014) 89–96
91
Glucocorticoids and Perioperative AF
C. Liu et al.
12,16,39,40,43]. Among these trials, 6 trials favor glucocorticoids [4,9,11,16,39,40]; whereas 3 trials reported there was no difference between two groups [10,12,43]. Nine out of thirty trials report their data on length of hospital stay as median and range or IQR. Two trails favor glucocorticoids [4,9]; whereas seven trials reported there was no difference between two groups [10– 12,16,39,40,43]. However, the largest trial involving 4494 participants reported that the mean time to discharge from the ICU was 34.2 h in the dexamethasone group versus 43.6 h in the placebo group (P < 0.001) and mean time of length of hospital stay in the dexamethasone group was 11.3 days versus 11.7 days in the placebo group (P = 0.009), favoring dexamethasone.
Effects of Glucocorticoid Prophylaxis on Adverse Event
Figure 3 Meta-regression using doses of glucocorticoid as a predictor of risk of atrial fibrillation in adult cardiac surgery (slope of the regression line and 95% CI).
Glucocorticoid prophylaxis was not associated with increase of all cause perioperative infection (RR 0.68, 95% CI, 0.58–0.78, n = 6512, P < 0.001). There was a trend that glucocorticoid prophylaxis might reduce mortality related to cardiac surgery (RR, 0.75, 95% CI, 0.52–1.08, n = 6902, P = 0.12).
Figure 4 The effect of glucocorticoid prophylaxis on length of ICU stay in adult cardiac surgery.
92
Cardiovascular Therapeutics 32 (2014) 89–96
ª 2014 John Wiley & Sons Ltd
Glucocorticoids and Perioperative AF
C. Liu et al.
Figure 5 The effect of glucocorticoid prophylaxis on length of hospital stay in adult cardiac surgery.
Sensitivity Analyses and Publication Bias After the analysis was restricted to trials with high quality design, using b-blockers, using POAF as primary clinical outcome, or participants underwent CABG, the effect of glucocorticoid prophylaxis on POAF remained unchanged (Table 1). There were a total of 68 RCTs involving glucocorticoid prophylaxis in adult cardiac surgery with high heterogeneity in their prespecified primary outcomes. Only 26 RCTs reported their data on the occurrence of POAF. Even in those RCTs that reported detailed data on POAF, only 5 RCTs used POAF as the primary outcome. The funnel plot suggested there was no publication bias in the trials used POAF for their primary outcome (Figure 6).
Discussion This meta-analysis showed that glucocorticoid prophylaxis significantly reduced the incidence of atrial fibrillation and length of
ª 2014 John Wiley & Sons Ltd
ICU and hospital stay. This benefit was not significantly different between low, medium and high doses of glucocorticoid. This significant clinical benefit obtained by glucocorticoid prophylaxis further supports the role inflammation in cardiac surgery and potential therapeutic role of anti-inflammatory agents in this setting. The systemic inflammatory response observed in patients undergoing cardiopulmonary bypass and cardiac surgery increases plasma levels of several acute phase proteins such as CRP and TNF are usually short-lived, but can be severe and prolonged in some cases [6,18,20,24,25]. The sustained increase in cytokine levels is associated with adverse outcomes such as cardiovascular events, and stroke and thromboembolism [47]. Glucocorticoids, as anti-inflammatory drugs, are widely used for the treatment of numerous inflammatory diseases. Twenty-two out of 42 RCTs in the present meta-analysis reported the effect of glucocorticoids on inflammatory response [3,5,6,11,14,16,18,20,24–27,32,35, 36,40–46]. All of them unanimously reported that glucocorticoid prophylaxis reduced inflammatory response with a reduction of
Cardiovascular Therapeutics 32 (2014) 89–96
93
Glucocorticoids and Perioperative AF
C. Liu et al.
Table 1 Sensitivity analysis with restriction
Subgroups Primary outcome was POAF b-blockers (≧50%) High quality design Patients with valvular diseases Patients with CABG
Study (n)
Patients (n)
RR with 95% CI
P-value
5
702
0.65 (0.53–0.80)
0.0001
8 13 3
5416 5811 161
0.77 (0.62–0.96) 0.71 (0.58–0.88) 0.39 (0.23–0.66)
0.02 0.002 0.0005
15
1581
0.75 (0.61–0.92)
0.006
Figure 6 The funnel plot for trials using risk of POAF as the primary end point.
CRP, TNF, IL-1, 6, 8 and complement components, which might contribute to the reduction of POAF. Theoretically, anti-inflammatory effects of glucocorticoids are dose-dependent. However, our results did not show that glucocorticoid prophylaxis with high dose was superior over low or medium dose of glucocorticoid prophylaxis in preventing POAF. It has been suggested that high doses of glucocorticoids may alter the phospholipid cell membrane, causing potassium flux across the cell membrane [48]. In humans, intravenous methylprednisolone alters the stimulation threshold of myocardial cells and the urinary excretion of both potassium and sodium. These changes might alter electrolyte shifts across myocardial cell membranes, resulting in cardiac arrhythmias. It is consistent with pervious reports that intravenous use of large doses of glucocorticoids was associated with cardiac arrhythmia [49,50]. In viewing of the fact that most of data are from trials using medium dose of glucocorticoids, more reliance should be on the medium dose of glucocorticoids when used in patients undergoing cardiac surgery. As immunosuppressive drugs, glucocorticoids may increase the risk of infection after surgery. However, our results showed that the infection rate in glucocorticoid group was dramatically lower than that in control group. Short course of use of glucocorticoid prophylaxis (< 72 h) in cardiac surgery did not increase the risk of perioperative infection, but reduced infection rate. Possible explanation is that glucocorticoid prophylaxis could significantly
94
Cardiovascular Therapeutics 32 (2014) 89–96
reduce respiratory failure in patients with cardiac surgery [9]. The beneficial effects obtained by glucocorticoid prophylaxis in post-cardiac arrhythmia, length of ICU and hospital stay, and perioperative infection may be translated into the positive effect on mortality. The majority of the studies (61.8%) did not use POAF as an outcome. Even for the trials that reported occurrence of POAF, only five trials used POAF as their primary outcome. It, inevitably, posed high risk for POAF being missed due to their study design. Therefore, we only funnel plotted the risk of POAF for the trials that used POAF as primary outcome to assess publication bias. The present meta-analysis has several limitations. First, publication bias may be a big concern because many of the studies that did not have POAF as an endpoint were excluded from the analysis. Moreover, the variable duration of follow-up and incomplete monitoring for POAF in some of the included studies may affect the accuracy of the incidence of POAF reported in the present meta-analysis. However, the efficacy of glucocorticoid prophylaxis on POAF remains unchanged when we restricted meta-analysis in the trials that used POAF as their primary outcome. Furthermore, small size trials with negative results other than positive results are difficult to be published, leading to overrepresentation of a drug’s benefit in systematic reviews. However, there are no significant differences in the direction and magnitude of glucocorticoids’ benefit after we restricted the analysis to trials with larger sample size (n ≥ 50) studies. Third, medications such as betablockers and amiodarone proven to be effective at preventing POAF may have been used in the trials. But, their use was seldom reported in the included studies. Nine included trials provided detailed data about beta-blocker use (Table S1). Only one trial clearly reported they did not use amiodarone [43], the data about amiodarone in other 41 trials were not available. Whether adding glucocorticoid on the top of beta-blocker or amiodarone would further reduce POAF incidence remains inconclusive. However, our limited data showed glucocorticoids might further reduce risk of POAF when we pooled the data from the trials that the percentage of beta-blocker use was larger than 50%. Finally, a wide range of doses, and different preparations of glucocorticoid used in the different studies make it harder to identify the ideal dosage and administration time of glucocorticoid in patients undergoing cardiac surgery.
Conclusions The results of present met-analysis suggests that both low-dose of and medium-dose of glucocorticoid could effectively reduce the risk of POAF without increasing the advent events in patients undergoing cardiac surgery. The beneficial effect on POAF is associated with a reduced length of ICU and hospital stay, and infection rate.
Funding Sources China Scholarship Council grant (201208130092).
Conflict of Interest The authors have no conflicts of interest to disclose.
ª 2014 John Wiley & Sons Ltd
Glucocorticoids and Perioperative AF
C. Liu et al.
References
with severe lv dysfunction undergoing elective off-pump coronary artery bypass surgery. Acta
1. Mathew JP, Fontes ML, Tudor IC, et al., Investigators of the Ischemia R, Education F,
Medica Iranica 2011;49:288–292. 14. Kilger E, Heyn J, Beiras-Fernandez A, Luchting
cardiopulmonary bypass. Crit Care Med 2001;29:2137–2142. 26. El Azab SR, Rosseel PM, de Lange JJ, Groeneveld AB, van Strik R, van Wijk EM,
Multicenter Study of Perioperative Ischemia
B, Weis F. Stress doses of hydrocortisone reduce
Scheffer GJ. Dexamethasone decreases the pro-
Research G. A multicenter risk index for atrial
systemic inflammatory response in patients
to anti-inflammatory cytokine ratio during
fibrillation after cardiac surgery. JAMA
undergoing cardiac surgery without
2004;291:1720–1729.
cardiopulmonary bypass. Minerva Anestesiol
2. McKeown P, Epstein AE, American College of Chest P. Future directions: American college of
2011;77:268–274. 15. Yasser Mohamed A, Elmistekawy E, El-Serogy
chest physicians guidelines for the prevention
H. Effects of dexamethasone on pulmonary and
and management of postoperative atrial
renal functions in patients undergoing cabg
fibrillation after cardiac surgery. Chest
with cardiopulmonary bypass. Semin Cardiothorac
2005;128:61S–64S. 3. Fillinger MP, Rassias AJ, Guyre PM, et al.
Vasc Anesth 2009;13:231–237. 16. Weis F, Beiras-Fernandez A, Schelling G, et al.
Glucocorticoid effects on the inflammatory and
Stress doses of hydrocortisone in high-risk
clinical responses to cardiac surgery. J
patients undergoing cardiac surgery: Effects on
Cardiothorac Vasc Anesth 2002;16:163–169.
interleukin-6 to interleukin-10 ratio and
4. Kilger E, Weis F, Briegel J, et al. Stress doses of hydrocortisone reduce severe systemic
cardiac surgery. Crit Care Med 2003;31:1068–1074.
reperfusion phenomena in cardiopulmonary
dexamethasone on the incidence of shivering and recovery in patients undergoing valve replacement surgery. Eg J Anaesth 2003;19:361– 370. 29. Halvorsen P, Raeder J, White PF, Almdahl SM,
revascularization procedures. Anesth Analg 2003;96:1578–1583. 30. Volk T, Schmutzler M, Engelhardt L, et al.
bypass. J Thorac Cardiovasc Surg 1991;102:515–
Effects of different steroid treatment on
525.
reperfusion-associated production of reactive
reaction occurring during cardiopulmonary
18. Engelman RM, Rousou JA, Flack JE III, Deaton
bypass: Effects on tnf-alpha, il-6, il-8, il-10.
DW, Kalfin R, Das DK. Influence of steroids on
Perfusion 2004;19:185–191.
complement and cytokine generation after
prednisolone on serum cytokine levels following
205–214. 28. Abd El-Hakeem EE, Zareh ZE. Effects of
dexamethasone on side effects after coronary
et al. Inhibition by dexamethasone of the
OA, Cengiz M. Effect of low-dose methyl
replacement surgery. Eg J Anaesth 2003;19:
Nordstrand K, Saatvedt K, Veel T. The effect of
early outcome in a risk group of patients after
6. Yilmaz M, Ener S, Akalin H, Sagdic K, Serdar
balance in patients undergoing valve
early outcome. Crit Care Med 2009;37:1685– 17. Jansen NJ, van Oeveren W, van den Broek L,
H. Methylprednisolone prevents inflammatory
Influence of dexamethasone on cytokine
1690.
inflammatory response syndrome and improve
5. Celik JB, Gormus N, Okesli S, Gormus ZI, Solak
cardiac surgery. Br J Anaesth 2002;88:496–501. 27. Abd El-Hakeem EE, El-Minshawy AM.
cardiopulmonary bypass. Ann Thorac Surg 1995;60:801–804. 19. Chaney MA, Nikolov MP, Blakeman B, Bakhos
oxygen species and arrhythmias during coronary surgery. Acta Anaesthesiol Scand 2003;47:667–674. 31. Loef BG, Henning RH, Epema AH, Rietman GW, van Oeveren W, Navis GJ, Ebels T. Effect of dexamethasone on perioperative renal
extracorporeal circulation. Perfusion 1999;
M, Slogoff S. Pulmonary effects of
function impairment during cardiac surgery
14:201–206.
methylprednisolone in patients undergoing
with cardiopulmonary bypass. Br J Anaesth
7. Baker WL, White CM, Kluger J, Denowitz A, Konecny CP, Coleman CI. Effect of perioperative corticosteroid use on the incidence of postcardiothoracic surgery atrial fibrillation and length of stay. Heart Rhythm 2007;4:461–468. 8. Marik PE, Fromm R. The efficacy and dosage
coronary artery bypass grafting and early tracheal extubation. Anesth Analg 1998;87:27– 33. 20. Tassani P, Richter JA, Barankay A, et al. Does high-dose methylprednisolone in aprotinin-treated patients attenuate the systemic
2004;93:793–798. 32. McBride WT, Allen S, Gormley SM, et al. Methylprednisolone favourably alters plasma and urinary cytokine homeostasis and subclinical renal injury at cardiac surgery. Cytokine 2004;27:81–89. 33. Oliver WC Jr, Nuttall GA, Orszulak TA, Bamlet
effect of corticosteroids for the prevention of
inflammatory response during coronary artery
atrial fibrillation after cardiac surgery: A
bypass grafting procedures? J Cardiothorac Vasc
WR, Abel MD, Ereth MH, Schaff HV.
systematic review. J Crit Care 2009;24:
Anesth 1999;13:165–172.
Hemofiltration but not steroids results in earlier
458–463. 9. Dieleman JM, Nierich AP, Rosseel PM, et al.
21. Wan S, LeClerc JL, Huynh CH, Schmartz D, DeSmet JM, Yim AP, Vincent JL. Does steroid
Intraoperative high-dose dexamethasone for
pretreatment increase endotoxin release during
cardiac surgery: A randomized controlled trial.
clinical cardiopulmonary bypass? J Thorac
JAMA 2012;308:1761–1767. 10. Abbaszadeh M, Khan ZH, Mehrani F,
Cardiovasc Surg 1999;117:1004–1008. 22. Yared JP, Starr NJ, Torres FK, et al. Effects of
Jahanmehr H. Perioperative intravenous
single dose, postinduction dexamethasone on
corticosteroids reduce incidence of atrial
recovery after cardiac surgery. Ann Thorac Surg
fibrillation following cardiac surgery: A
2000;69:1420–1424.
randomized study. Rev Bras Cir Cardiovasc
tracheal extubation following cardiopulmonary bypass: A prospective, randomized double-blind trial. Anesthesiology 2004;101:327–339. 34. Bingol H, Cingoz F, Balkan A, Kilic S, Bolcal C, Demirkilic U, Tatar H. The effect of oral prednisolone with chronic obstructive pulmonary disease undergoing coronary artery bypass surgery. J Card Surg 2005;20:252–256. 35. Prasongsukarn K, Abel JG, Jamieson WR,
23. Chaney MA, Durazo-Arvizu RA, Nikolov MP,
Cheung A, Russell JA, Walley KR, Lichtenstein
Blakeman BP, Bakhos M. Methylprednisolone
SV. The effects of steroids on the occurrence of
does not benefit patients undergoing coronary
postoperative atrial fibrillation after coronary
et al. Steroids and statins: An old and a new
artery bypass grafting and early tracheal
artery bypass grafting surgery: A prospective
anti-inflammatory strategy compared. Perfusion
extubation. J Thorac Cardiovasc Surg
randomized trial. J Thorac Cardiovasc Surg
2011;26:31–37.
2001;121:561–569.
2012;27:18–23. 11. Vukovic PM, Maravic-Stojkovic VR, Peric MS,
12. Murphy GS, Sherwani SS, Szokol JW, et al.
24. Schurr UP, Zund G, Hoerstrup SP, Grunenfelder
2005;130:93–98. 36. Rubens FD, Nathan H, Labow R, et al. Effects of
Small-dose dexamethasone improves quality of
J, Maly FE, Vogt PR, Turina MI. Preoperative
methylprednisolone and a biocompatible
recovery scores after elective cardiac surgery: A
administration of steroids: Influence on
copolymer circuit on blood activation during
randomized, double-blind, placebo-controlled
adhesion molecules and cytokines after
cardiopulmonary bypass. Ann Thorac Surg
study. J Cardiothorac Vasc Anesth 2011;25:950–
cardiopulmonary bypass. Ann Thorac Surg
2005;79:655–665.
960. 13. Mirhosseini SJ, Forouzannia SK, Sayegh AH,
2001;72:1316–1320. 25. Volk T, Schmutzler M, Engelhardt L, Docke
37. Enc Y, Karaca P, Ayoglu U, Camur G, Kurc E, Cicek S. The acute cardioprotective effect of
Sanatkar M. Effect of prophylactic low dose of
WD, Volk HD, Konertz W, Kox WJ. Influence
glucocorticoid in myocardial
methylprednisolone on postoperative new atrial
of aminosteroid and glucocorticoid treatment on
ischemia-reperfusion injury occurring during
fibrillation and early complications in patients
inflammation and immune function during
ª 2014 John Wiley & Sons Ltd
Cardiovascular Therapeutics 32 (2014) 89–96
95
Glucocorticoids and Perioperative AF
C. Liu et al.
cardiopulmonary bypass. Heart Vessels
methylprednisolone in patients undergoing
2006;21:152–156.
cardiac surgery. Ann Thorac Surg 2007;84:110–
38. Sano T, Morita S, Masuda M, Yasui H. Minor infection encouraged by steroid administration
118; discussion 118-119 43. Yared JP, Bakri MH, Erzurum SC, et al. Effect
during cardiac surgery. Asian Cardiovasc Thorac
of dexamethasone on atrial fibrillation after
Ann 2006;14:505–510.
cardiac surgery: Prospective, randomized,
39. Weis F, Kilger E, Roozendaal B, et al. Stress doses of hydrocortisone reduce chronic stress symptoms and improve health-related quality of
double-blind, placebo-controlled trial. J Cardiothorac Vasc Anesth 2007;21:68–75. 44. Kilickan L, Yumuk Z, Bayindir O. The effect of
derivative to decrease possible neuronal damage. J Card Surg 2009;24:397–403. 47. Mendall MA, Patel P, Asante M, et al. Relation of serum cytokine concentrations to cardiovascular risk factors and coronary heart disease. Heart 1997;78:273–277. 48. Barnes PJ. How corticosteroids control inflammation: Quintiles prize lecture 2005. Br J Pharmacol 2006;148:245–254.
life in high-risk patients after cardiac surgery: A
combined preinduction thoracic epidural
randomized study. J Thorac Cardiovasc Surg
anaesthesia and glucocorticoid administration
supraventricular tachycardia after intravenous
2006;131:277–282.
on perioperative interleukin-10 levels and
pulse methylprednisolone therapy, with a brief
hyperglycemia. A randomized controlled trial. J
review of the literature. Rheumatol Int
40. Whitlock RP, Young E, Noora J, Farrokhyar F, Blackall M, Teoh KH. Pulse low dose steroids attenuate post-cardiopulmonary bypass sirs; sirs i. J Surg Res 2006;132:188–194.
Cardiovasc Surg 2008;49:87–93. 45. Sobieski MA II, Graham JD, Pappas PS, Tatooles
49. Kumari R, Uppal SS. First report of
2005;26:70–73. 50. Moretti R, Torre P, Antonello RM, Zorzon M,
AJ, Slaughter MS. Reducing the effects of the
Cazzato G. Recurrent atrial fibrillation
systemic inflammatory response to
associated with pulse administration of high
Corticosteroids for the prevention of atrial
cardiopulmonary bypass: Can single dose
doses of methylprednysolone: A possible
fibrillation after cardiac surgery: A randomized
steroids blunt systemic inflammatory response
prophylactic treatment. Eur J Neurol 2000;7:130.
controlled trial. JAMA 2007;297:1562–1567.
syndrome? ASAIO J 2008;54:203–206.
41. Halonen J, Halonen P, Jarvinen O, et al.
42. Liakopoulos OJ, Schmitto JD, Kazmaier S,
46. Demir T, Demir H, Tansel T, et al. Influence of
Brauer A, Quintel M, Schoendube FA, Dorge H.
methylprednisolone on levels of neuron-specific
Cardiopulmonary and systemic effects of
enolase in cardiac surgery: A corticosteroid
Supporting Information Additional Supporting Information may be found in the online version of this article: Table S1 Clinical characteristics of included studies.
96
Cardiovascular Therapeutics 32 (2014) 89–96
ª 2014 John Wiley & Sons Ltd
