胺碘酮和利多卡因对开胸心脏手术中再灌注性室颤预防效果的随机对照试验荟萃分析

Efficacy of amiodarone and lidocaine for preventing ventricular fibrillation after aortic cross-clamp release in open heart surgery: a meta-analysis of randomized controlled trials 胺碘酮和利多卡因对开胸心脏手术中再灌注性室颤预防效果的随机对照试验荟萃分析 1 2 first-author 3 4 5 6 1 2 1 2 corresp ‡Corresponding Author ‡Corresponding Author †E-mail:xiangmx@zju.edu.cn †E-mail:xiangmx@zju.edu.cn xiangmx@zju.edu.cn Department of Cardiology, the Second Affiliated Hospital, School of Medicine, Zhejiang University, Hangzhou 310009, China Department of Cardiology, the Second Affiliated Hospital, School of Medicine, Zhejiang University, Hangzhou 310009, China Cardiovascular Key Lab of Zhejiang Province, the Second Affiliated Hospital, School of Medicine, Zhejiang University, Hangzhou 310009, China Cardiovascular Key Lab of Zhejiang Province, the Second Affiliated Hospital, School of Medicine, Zhejiang University, Hangzhou 310009, China Department of Anesthesiology, the First Affiliated Hospital, School of Medicine, Zhejiang University, Hangzhou 310003, China Department of Anesthesiology, the First Affiliated Hospital, School of Medicine, Zhejiang University, Hangzhou 310003, China Cardiology Division, the First Affiliated Hospital of Nanjing Medical University, Nanjing 210029, China Cardiology Division, the First Affiliated Hospital of Nanjing Medical University, Nanjing 210029, China Department of Respiratory, Tinhu People's Hospital, Yancheng 224001, China Department of Respiratory, Tinhu People's Hospital, Yancheng 224001, China Department of Biology, Georgia Southern University, Statesboro 30460, USA Department of Biology, Georgia Southern University, Statesboro 30460, USA Objective: 2 The relative preventative efficacy of amiodarone and lidocaine for ventricular fibrillation (VF) after release of an aortic cross-clamp (ACC) during open heart surgery has not been determined. This meta-analysis was designed to systematically evaluate the influence of amiodarone, lidocaine, or placebo on the incidence of VF after ACC. Methods: 2 Prospective randomized controlled trials (RCTs) that compared the VF-preventative effects of amiodarone with lidocaine, or amiodarone or lidocaine with placebo were included. PubMed, EMBASE, and the Cochrane Library were searched for relevant RCTs. Fixed or randomized effect models were applied according to the heterogeneity of the data from the selected studies. Results: 2 We included eight RCTs in the analysis. Pooled results suggested that the preventative effects of amiodarone and lidocaine were comparable (relative risk (RR)=1.12, 95% confidence interval (CI): 0.70 to 1.80, P =0.63), but both were superior to the placebo (amiodarone, RR=0.71, 95% CI: 0.51 to 1.00, P =0.05; lidocaine, RR=0.63, 95% CI: 0.46 to 0.88, P =0.006). The percentage of patients requiring electric defibrillation counter shocks (DCSs) did not differ significantly among patients administered amiodarone (RR=0.21, 95% CI: 0.04 to 1.19, P =0.08), lidocaine (RR=2.44, 95% CI: 0.13 to 44.02, P =0.55), or the placebo (RR=0.56, 95% CI: 0.25 to 1.25, P =0.16). Conclusions: 2 Amiodarone and lidocaine are comparably effective in preventing VF after ACC, but the percentage of patients who subsequently require DCSs does not differ among those administered amiodarone, lidocaine, or placebo. 目的 2 系统评估胺碘酮、利多卡因或安慰剂对心脏手术中主动脉结扎松解（ACC）后再灌注性室颤（VF）发生的影响。 创新点 2 胺碘酮、利多卡因对于预防ACC后再灌注VF的效果，目前相关的随机对照试验（RCTs）并无统一的结论，本研究综合之前RCTs进行荟萃分析。 方法 2 遵循PRISMA和Cochrane系统评估手册对PubMed、EMBASE及Cochrane进行文献检索 （图1），荟萃分析符合要求的RCTs。 结论 2 当前的证据表明胺碘酮和利多卡因在开胸心脏手术中预防再灌注VF的效果两者之间并无显著差异，但均明显优于安慰剂；在随后需要电除颤的患者的比例上，胺碘酮、利多卡因及安慰剂三者间无统计学差异。 Amiodarone Lidocaine Ventricular fibrillation Open heart surgery 胺碘酮 利多卡因 再灌注室颤 心脏手术 Introduction 1 Despite significant advancements in the treatment of cardiovascular diseases (Liu et al., 2010 2010 ), open heart surgery remains one of the most important interventions for patients with severe cardiovascular diseases, such as those with multiple-vessel coronary heart diseases (CHD) (Verrier and Mack, 2015 2015 ; Papakonstantinou et al., 2017 2017 ), severe valvular heart diseases (Geissler et al., 2009 2009 ), and complicated congenital heart diseases (Mery and Kane, 2017 2017 ). With improvements in perioperative care and cardiopulmonary bypass (CPB) techniques, the indications for cardiac surgery have been expanded and many patients who were previously deemed to be contraindicated or intolerant to cardiac surgeries are now considered suitable for open heart surgery (Marathe and Talwar, 2015 2015 ). However, patients who undergo cardiac surgery are still at risk of fatal complications associated with both the surgical procedures and CPB, such as major bleeding, infection, and most importantly, lethal arrhythmias, such as ventricular fibrillation (VF) (Shake et al., 2013 2013 ; Zhan et al., 2015 2015 ). Pathophysiologically, VF is associated with myocardial ischemia and reperfusion injury, and subsequent inflammatory responses, oxidative stress, and electrical instability (Yeh et al., 2013 2013 ; Patil et al., 2015 2015 ). VF occurs most frequently after aortic cross-clamp (ACC) release during the coronary artery bypass graft (CABG) procedure, when myocardium reperfusion is initiated (Leeuwenburgh et al., 2008 2008 ). The incidence of VF after ACC release is reported to vary with the experience of the surgeon and the category of surgical procedures, ranging from 45% to 90% in patients undergoing CABG (Leeuwenburgh et al., 2008 2008 ; Baravelli et al., 2010 2010 ). When VF occurs during cardiac surgery, the most effective treatment strategy is electric defibrillation counter shocks (DCSs). However, the success rate of DCSs for patients with VF during open heart cardiac surgery is not satisfactory. Moreover, even in patients for whom VF was successfully terminated after DCSs, the procedure is associated with potential secondary damage to the myocardium, possibly leading to cardiac dysfunction and recurrent lethal arrhythmia (Doherty et al., 1979 1979 ; Yamaguchi et al., 2002 2002 ). Therefore, prophylactic measures capable of reducing the risk of VF after ACC release in patients undergoing cardiac surgery are urgently sought to improve the prognosis of these patients. Perioperative administration of conventional antiarrhythmic medications has been suggested for effective prevention of VF after ACC release in patients undergoing cardiac surgery (Hippelainen et al., 1994 1994 ; Schluter et al., 2001 2001 ). The capacity of two conventional antiarrhythmic medications, amiodarone and lidocaine, to prevent VF during cardiac surgery has been assessed clinically, but the results are contradictory. Two early studies concluded that a cardioplegic solution containing lidocaine can significantly reduce VF incidence by 22%–42% in patients undergoing cardiac surgery (Fiore et al., 1990 1990 ; Baraka et al., 1993 1993 ). Moreover, another study indicated that administration of lidocaine with a perfusion pump before ACC release reduced VF incidence from 70% to 11% (Baraka et al., 2000 2000 ). As for amiodarone, subsequent studies demonstrated that this medication may achieve comparable (Samantaray et al., 2010 2010 ) or even greater (Dorian et al., 2002 2002 ) preventative effects against VF in patients who are at risk for development of VF and pulseless ventricular tachycardia during cardiac surgery. In contrast, several other studies reported that neither amiodarone nor lidocaine effectively prevented VF in cardiac surgeries. A randomized controlled trial (RCT) conducted by Ayoub et al. ( 2009 2009 ) concluded that VF incidence did not differ significantly between patients who received amiodarone and those who received placebo. Furthermore, a subsequent three-armed RCT reported no significant difference in VF incidence after ACC release in patients undergoing cardiac surgery who were administered amiodarone, lidocaine, or placebo (Ghavidel et al., 2013 2013 ). Many factors have been proposed to explain these inconsistencies in clinical trial results, including their inadequate statistical power produced by small numbers of patients. Therefore, in this study, we performed a meta-analysis of head-to-head RCTs to compare the capacity of amiodarone, lidocaine, and placebo to prevent VF after ACC release in patients undergoing cardiac surgery. Materials and methods 1 Literature search 2 Our study followed the recommendations of the PRISMA 2009 checklist (Moher et al., 2009 2009 ) and the Cochrane Handbook for Systematic Reviews of Interventions 5.0.2. Electronic databases (PubMed, EMBASE, and Cochrane Central Register of Controlled Trials) were searched for relevant RCTs with the keywords amiodarone or lidocaine, paired with ventricular fibrillation or reperfusion. The language was restricted to English and the final search was performed on August 25th, 2017. Moreover, the reference lists from the identified articles were manually searched for additional relevant studies. Two investigators (YZ, QG) independently reviewed the literature and articles fulfilling the following criteria were selected: (1) designed as an RCT; (2) included adult patients undergoing cardiac surgery requiring ACC; (3) patients were assigned to either 2 or 3 of the following interventional groups: amiodarone, lidocaine, or placebo; (4) reported outcome data included VF incidence or percentage of patients requiring electric DCSs. Reviews, case reports, retrospective studies, observational studies, or studies with unavailable outcome data were excluded. Data extraction and quality assessment 2 We extracted data regarding patient demographics (age, gender), type of surgery, duration of ACC, and duration of CPB. Where possible we used intention-to-treat (ITT) data, but where this was unavailable we extracted data from tables or graphs in each article or retrieved data directly from authors. The quality of all selected studies was evaluated based on the Cochrane Back Review Group 12-item scale (Furlan et al., 2015 2015 ). Two reviewers (YZ, QG) independently extracted data and evaluated the quality of the RCTs, and any disagreements were resolved by consulting a third investigator (MX). Statistical analysis 2 Dichotomous variables were assessed according to relative risk (RR) and 95% confidence interval (CI). The Cochrane’s Q test and I 2 test were used to measure the statistical heterogeneity. I 2 values of 25%, 50%, and 75% were considered to indicate low, moderate, and high statistical heterogeneity, respectively. Random effect models were used to pool the results as these can incorporate the inherited heterogeneity among the studies. Subgroup analyses and meta-regression were performed to evaluate potential sources of heterogeneity if sufficient numbers of interventional groups were included. Publication bias among the included studies was analyzed by a funnel plot and Egger’s regression test. All analyses were conducted with RevMan software (Review Manager, Version 5.3, the Cochrane Collaboration, Oxford, UK) and STATA software 12.0 (StataCorp, College Station, Texas, USA). Results 1 Literature review 2 The initial search retrieved 4465 relevant citations (PubMed 1493, EMBASE 2634, and the Cochrane’s Library 338), and finally 8 RCTs were included in our meta-analysis. The literature searching procedure is described in Fig. 1 1 . Quality assessment and characteristics of included studies 2 Overall, 8 RCTs were selected, which included a total of 839 patients: 249 treated with amiodarone, 288 treated with lidocaine, and 302 with placebo. All included studies were RCTs and, according to the item standard quality evaluation, five were classified as high-quality and the other three as moderate quality studies (Table 1 1 ). Among these studies, four were three-arm trials (Ayoub et al., 2009 2009 ; Mauermann et al., 2012 2012 ; Ghavidel et al., 2013 2013 ; Yilmaz et al., 2014 2014 ) of amiodarone versus lidocaine versus placebo, three compared lidocaine with placebo (Kirlangitis et al., 1990 1990 ; Baraka et al., 2000 2000 ; Vaziri et al., 2010 2010 ), and one compared amiodarone with placebo (Samantaray et al., 2010 2010 ). In all studies, all patients were scheduled to undergo elective surgery and were age-, gender-, and operative condition-matched (Table 2 2 ). Only if the sequencing method was explicitly introduced was a study given a ‘‘Yes’’; sequence generated by ‘‘Dates of Admission’’ or ‘‘Patient’s Number’’ received a ‘‘No’’; Drop-out rates <20% were given a ‘‘Yes’’, otherwise ‘‘No’’; ITT: intention-to-treat; only if all randomized participants were analyzed in the group to which they were allocated was a study given a ‘‘Yes’’; Studies with more than 7 items were scored as ‘‘High’’; those with more than 4, but no more than 7 items were scored as ‘‘Moderate’’; those with no more than 4 items were scored as ‘‘Low’’ Study Randomized adequately a a Allocation concealed Patient blinded Care provider blinded Outcome assessor blinded Acceptable drop-out rate b b ITT analysis c c Kirlangitis et al., 1990 1990 No No Yes Unsure Unsure Yes No Baraka et al., 2000 2000 No No Yes Unsure Unsure Yes Yes Ayoub et al., 2009 2009 No No Yes Unsure Unsure Yes Yes Samantaray et al., 2010 2010 No Yes Yes Unsure Unsure Yes Yes Vaziri et al., 2010 2010 No No Yes Unsure Unsure Yes No Mauermann et al., 2012 2012 No No Yes Unsure Unsure Yes Yes Ghavidel et al., 2013 2013 No No Yes Unsure Unsure Yes Yes Yilmaz et al., 2014 2014 No No Yes Unsure No Yes Yes 8 8 Study Avoided selective reporting Similar baseline Similar or avoided cofactor Patient compliance Similar timing Quality d d 8 Kirlangitis et al., 1990 1990 Yes Yes Yes Yes Yes Moderate Baraka et al., 2000 2000 Yes Yes Yes Yes Yes High Ayoub et al., 2009 2009 Yes Yes Yes Yes Yes High Samantaray et al., 2010 2010 Yes Yes Yes Yes Yes High Vaziri et al., 2010 2010 Yes Yes Yes Yes Yes Moderate Mauermann et al., 2012 2012 Yes Yes Yes Yes Yes High Ghavidel et al., 2013 2013 Yes Yes Yes Yes Yes High Yilmaz et al., 2014 2014 Yes Yes Yes Yes No Moderate LVEF, left ventricular ejection fraction; NA, not available; None, no amiodarone or lidocaine or placebo in this study; CABG, coronary artery bypass graft; Valve, valve surgery; Myectomy, septal myectomy; ACC, aortic cross-clamp; CPB, cardiopulmonary bypass; RVF, reperfusion ventricular fibrillation; DCSs, defibrillation counter shocks. Data are presented as mean±standard deviation (SD). From left to right, the data of amiodarone, lidocaine, and placebo, respectively Study Age (year) * * Gender, male (%) 2 LVEF (%) Type of surgery Kirlangitis et al., 1990 1990   None/(57.0±2.7)/(63.0±2.3) None/81.8/81.8 2   None/(54.0±2.9)/(61±2.4) CABG Baraka et al., 1993 1993   None/(60.8±10.8)/(58.5±12.7) None/76.5/82.4 2   None/>60/>60 CABG Ayoub et al., 2009 2009   (63±9)/(64±9)/(65±10) 90/92.5/90 2   >35/>35/>35 CABG Samantaray et al., 2010 2010   (47.2±6.6)/none/(50.0±6.0) 64.7/none/70.6 2   >35/none/>35 CABG Vaziri et al., 2010 2010   None/(56.7±7.5)/(60.8±10.5) 2   None/>50/>50 CABG Mauermann et al., 2012 2012   (63.3±13.6)/(62.7±13.9)/(63.6±13.0) 74/81/82 2   (62.7±10.8)/(62.7±11.5)/(62.8±11.7) CABG, Valve, and Myectomy Ghavidel et al., 2013 2013   (58.06±10.47)/(60.64±15.62)/(57.43±10.97) 78.4/80/86.4 2   (42.65±6.80)/(43.64±6.93)/(43.52±7.36) CABG Yilmaz et al., 2014 2014   (57.2±7.9)/(61.6±8.6)/(59.7±9.8) 81/76/80 2   (53.2±10.3)/(52.8±9.0)/(52.5±9.0) CABG 6 6 Study pH 2 ACC time (min) 2 CPB time (min) 6 Kirlangitis et al., 1990 1990   None/(7.41±0.02)/(7.49±0.02) 2 None/(56±4.6)/(54±7.3) 2   None/(119±12.5)/(103±10.5) Baraka et al., 1993 1993 2 None/(43.8±15.3)/(40.6±15.7) 2   None/(70.1±22.9)/(64.6±29.3) Ayoub et al., 2009 2009   (7.40±0.05)/(7.41±0.03)/(7.41±0.04) 2 (42.0±20.0)/(44.0±22.0)/(35.0±11.0) 2   (66.0±32.0)/(67.0±33.0)/(64.0±29.0) Samantaray et al., 2010 2010 2 (49.4±12.3)/none/(48.8±15.2) 2   (76.1±16.5)/none/(74.2±18.7) Vaziri et al., 2010 2010   None/(7.40±0.05)/(7.40±0.06) 2 None/(48.6±9.1)/(47.2±6.6) 2 Mauermann et al., 2012 2012 2 (47.4±32.1)/(46.5±56.1)/(53.3±36.8) 2   (70.8±64.0)/(74.3±40.0)/(78.0±48.7) Ghavidel et al., 2013 2013   (7.35±0.07)/(7.34±0.06)/(7.36±0.06) 2 (38.2±19.6)/(35.6±12.6)/(34.9±14.0) 2   (72.8±29.2)/(72.1±21.2)/(65.1±29.7) Yilmaz et al., 2014 2014   (7.40±0.04)/(7.50±0.51)/(7.40±0.06) 2 (67.6±19.7)/(64.1±18.9)/(63.2±8.8) 2   (104.1±31.3)/(113.6±27.8)/(114.4±27.6) 6 6 Study Core temperature (time of cross clamp release) (°C) 2 Timing of electrical defibrillation 2   Timing of the medication given before ACC release (min) 6 Kirlangitis et al., 1990 1990 24–28 2 RVF persist untreated for 1 min 2   None/5/5 Baraka et al., 1993 1993 29 2 Lidocaine100 mg+DCSs 2   None/2/2 Ayoub et al., 2009 2009 29 2 DCSs immediately after RVF 2   2/2/2 Samantaray et al., 2010 2010 30 2 DCSs immediately after RVF 2   3/none/3 Vaziri et al., 2010 2010 30–32 2 RVF persist untreated for 2 min 2   None/5/5 Mauermann et al., 2012 2012 32 2 DCSs immediately after RVF 2   3/3/3 Ghavidel et al., 2013 2013 34 2 DCSs immediately after RVF 2   3/3/3 Yilmaz et al., 2014 2014 34 2 RVF persist untreated for 2 min 2   15/2/NA 6 6 Study Dose of the medication given 2 Potassium (mEq/L) (during cross clamp) 2 Cardioplegia solution 6 Kirlangitis et al., 1990 1990   None/(2 mg/kg)/constant volume 2 2   Crystalloid hyperkalemic cardioplegia Baraka et al., 1993 1993   None/100 mg/constant volume 2 None/(4.6±0.5)/(4.7±0.5) 2   Crystalloid hyperkalemic cardioplegia Ayoub et al., 2009 2009   150 mg/100 mg/constant volume 2 (4.7±1.0)/(4.8±0.5)/(4.9±0.5) 2   Crystalloid hyperkalemic cardioplegia Samantaray et al., 2010 2010   150 mg/none/constant volume 2 (4.4±0.3)/none/(4.5±0.5) 2   Crystalloid hyperkalemic cardioplegia Vaziri et al., 2010 2010   None/(1.5 mg/kg)/constant volume 2 None/(4.30±0.67)/(4.40±0.89) 2 Mauermann et al., 2012 2012   300 mg/(1.5 mg/kg)/constant volume 2 2   Crystalloid hyperkalemic cardioplegia Ghavidel et al., 2013 2013   150 mg/100 mg/constant volume 2 (4.10±0.44)/(4.11±0.42)/(4.24±0.47) 2   Retrograde St. Thomas solution Yilmaz et al., 2014 2014   300 mg/(1.5 mg/kg)/constant volume 2 (5.00±0.63)/(4.80±0.61)/(4.40±0.89) 2   Retrograde St. Thomas solution and crystalloid hyperkalemic cardioplegia Effects of amiodarone and lidocaine on VF incidence after release of ACC 2 Four of the eight trials, including a total of 466 patients, directly compared the efficacy of amiodarone and lidocaine. The rate of VF after release of ACC did not differ significantly between patients treated with amiodarone and lidocaine (RR=1.12, 95% CI: 0.70 to 1.80, P =0.63; Fig. 2a 2a ). Moderate heterogeneity was observed ( I 2 =58%, P =0.07). Analysis of the pooled results of five RCTs including 498 patients revealed that amiodarone was associated with a lower risk of VF than placebo (RR=0.71, 95% CI: 0.51 to 1.00, P =0.05; Fig. 2b 2b ) with moderate heterogeneity ( I 2 =48%, P =0.11). Similarly, by pooling the results of seven RCTs including 473 patients, we observed that lidocaine was also significantly associated with lower VF risk (RR=0.63, 95% CI: 0.46 to 0.88, P =0.006; Fig. 2c 2c ) than placebo with moderate heterogeneity ( I 2 =54%, P =0.04). Due to the limited number of studies, subsequent subgroup and meta-regression analyses were not performed. The rate of VF after release of ACC did not differ significantly between patients undergoing open heart surgery who were treated with amiodarone or lidocaine (a); amiodarone (b) and lidocaine (c) were associated with a lower risk of VF than placebo Publication bias was difficult to estimate due to the small number of comparisons included. Effects of amiodarone and lidocaine on rate of DCSs after release of ACC 2 Three of the eight trials including a total of 214 patients compared the percentage of patients requiring DCSs for VF. Our results showed that this rate did not differ significantly between amiodarone and placebo groups (RR=0.56, 95% CI: 0.25 to 1.25, P =0.16; Fig. 3a 3a ) with moderate heterogeneity ( I 2 =51%, P =0.13). Similarly, the percentage of patients requiring DCSs for VF did not differ significantly between patients receiving lidocaine and placebo (3 trials, 214 patients; RR=0.21, 95% CI: 0.04 to 1.19, P =0.08; Fig. 3b 3b ; I 2 =71%, P =0.03) and those receiving amiodarone and lidocaine (2 trials, 180 patients; RR=2.44, 95% CI: 0.13 to 44.02, P =0.55; Fig. 3c 3c ). However, significant heterogeneity was detected ( I 2 =84%, P =0.01). Subsequent subgroup and meta-regression analyses were not performed due to the small number of studies. Publication bias was difficult to estimate due to the small number of comparisons included. The rate did not differ significantly between amiodarone and placebo groups (a); the percentage of patients requiring DCSs for VF did not differ significantly between patients receiving lidocaine and placebo (b) and those receiving amiodarone and lidocaine (c) Discussion 1 By integrating all available RCTs, the results of our meta-analysis indicate that amiodarone and lidocaine confer comparable preventative efficacy for VF incidence after ACC release in patients undergoing open heart surgery, and that the effects of each were significant compared with placebos. However, when the percentage of patients subsequently requiring DCSs for reperfusion VF was applied as the measure, no significant effects among amiodarone, lidocaine, and placebo were detected. Overall, the included studies represented a small sample size, and moderate to significant heterogeneity was detected among the included RCTs. Based on the results, our study indicated that perioperative administrations of amiodarone and lidocaine confer similar preventative efficacy for VF incidence after ACC release in patients undergoing open heart surgery. These results may highlight the potential use of conventional antiarrhythmic medications for the prevention of VF during cardiac surgery procedures. Further studies with larger numbers of participants are needed to confirm our results, and to determine the protocol and regimens for the perioperative administrations of amiodarone and lidocaine during open heart surgery. The incidence of VF following ACC release during cardiac surgery is relatively high and is associated with poor prognosis in patients undergoing cardiac surgery. Previous studies indicated that medications administered to prevent arrhythmia may also effectively prevent VF during cardiac surgery (Wenger et al., 1984 1984 ; Kirlangitis et al., 1990 1990 ; Komori et al., 1999 1999 ; Ayoub et al., 2009 2009 ). Amiodarone and lidocaine are the most commonly used drugs for the prevention of heart arrhythmia (Wyman et al., 2004 2004 ; Nayeem Ul et al., 2013 2013 ; Onk and Erkut, 2015 2015 ). However, studies directly comparing the effects of amiodarone and lidocaine on VF following ACC release during cardiac surgery have rarely been reported. Our meta-analysis revealed that the capacity of amiodarone and lidocaine to prevent reperfusion VF was similar, consistent with previous RCTs (Mauermann et al., 2012 2012 ; Ghavidel et al., 2013 2013 ). However, some studies raised the possibility that lidocaine may more effectively prevent VF following ACC release in patients undergoing cardiac surgery (Ayoub e t al ., 2009 2009 ). Amiodarone acts to stabilize the membrane of cardiomyocytes and was widely used for the prevention and treatment of atrial and ventricular arrhythmia, including VF (Levine et al., 1996 1996 ; Petrovic et al., 1998 1998 ; Somberg et al., 2002 2002 ). Regarding the potential mechanisms by which amiodarone may prevent VF during cardiac surgery, some recently published experimental studies may provide some novel evidence. In a swine model of hemorrhage-induced VF, resuscitation combined with amiodarone and vasopressin after hemorrhagic circulatory arrest resulted in better 3-h survival, better preservation of hemodynamic parameters, and less myocardial injury than resuscitation with vasopressin only, indicating the efficacy of amiodarone for termination of VF (Zoerner and Semenas, 2014 2014 ). This observation is consistent with previous studies which showed that administration of amiodarone may be associated with a lower long-duration ventricular defibrillation threshold (Wu et al., 2011 2011 ; Chevalier et al., 2012 2012 ), suggesting a beneficial effect of amiodarone on malignant arrhythmias. Similarly, lidocaine was also recommended as an antiarrhythmic drug for recurrent VF (American Heart Association, 1992 1992 ; 2000 2000 ). However, few experimental studies have investigated the mechanism by which lidocaine may reduce VF incidence, and the effect of this drug on VF in animal models during cardiac surgery has rarely been reported. The mechanisms mediating lidocaine’s effect on the risk of VF after ACC release in patients undergoing open heart surgery require further investigation. In this study, we observed no significant differences in the effects of amiodarone, lidocaine, or placebo on the ratio of patients who subsequently required DCSs for the conversion of VF following cardiac surgery. This observation is inconsistent with the observed differences in VF incidence, perhaps as far fewer studies reported the DCS outcome, and the statistic power of the pooled analyses therefore was more limited. Our conclusions are limited by the scope of this study, which should be considered when interpreting the results. Firstly, we included the results of only 8 RCTs including a total of 839 patients. The small number of RCTs included and the small sample size of each included study restricted further analyses of the other factors that may influence the outcomes, such as patient demographic factors, surgical categories, and patient comorbidities. Obviously, large-scale RCTs with adequate statistic power are required to confirm our results. Secondly, the quality of the included studies was moderate, which led to the moderate grade of our meta-analyses. Thirdly, the results of our meta-analysis are based on the pooled results of crude RRs, which may be biased by potential confounding factors that were not balanced between groups during randomization. However, as a meta-analysis of RCTs, the chances of potential bias caused by residual confounding factors are low when compared with meta-analyses of observational studies. In addition, the optimal protocols and dosages for the administration of amiodarone and lidocaine during the perioperative periods remain to be determined in future studies. Lastly, publication bias among the included studies was difficult to estimate since only a small number of studies were available. Publication bias may exist since we extracted only English language articles. Conclusions 1 In conclusion, the current evidence from head-to-head RCTs suggests that amiodarone and lidocaine have comparable preventative efficacy for VF after ACC, but the ratio of patients who subsequently require DCSs does not differ between amiodarone, lidocaine, and placebo groups. Further studies with a larger number of participants will be required to confirm our results. References American Heart Association Guidelines for cardiopulmonary resuscitation and emergency cardiac care . JAMA , 1992 . 268 ( 16 ): 2171 - 2183 . 1404767 . Guidelines for cardiopulmonary resuscitation and emergency cardiac care American Heart Association Guidelines 2000 for cardiopulmonary resuscitation and emergency cardiovascular care. Part 6: advanced cardiovascular life support: section 5: pharmacology I: agents for arrhythmias . Circulation , 2000 . 102 ( 8 Suppl. ): I112 - I128 . 10966669 . Guidelines 2000 for cardiopulmonary resuscitation and emergency cardiovascular care. Part 6: advanced cardiovascular life support: section 5: pharmacology I: agents for arrhythmias CM Ayoub , PM Sfeir , P Bou-Khalil , 等 . Prophylactic amiodarone versus lidocaine for prevention of reperfusion ventricular fibrillation after release of aortic cross-clamp . Eur J Anaesthesiol , 2009 . 26 ( 12 ): 1056 - 1060 . 19809326 DOI: 10.1097/EJA.0b013e32832f0dfb http://doi.org/10.1097/EJA.0b013e32832f0dfb . Prophylactic amiodarone versus lidocaine for prevention of reperfusion ventricular fibrillation after release of aortic cross-clamp A Baraka , N Hirt , A Dabbous , 等 . Lidocaine cardioplegia for prevention of reperfusion ventricular fibrillation . Ann Thorac Surg , 1993 . 55 ( 6 ): 1529 - 1533 . 8512407 DOI: 10.1016/0003-4975(93)91104-U http://doi.org/10.1016/0003-4975(93)91104-U . Lidocaine cardioplegia for prevention of reperfusion ventricular fibrillation A Baraka , N Kawkabani , A Dabbous , 等 . Lidocaine for prevention of reperfusion ventricular fibrillation after release of aortic cross-clamping . J Cardiothor Vas An , 2000 . 14 ( 5 ): 531 - 533 . DOI: 10.1053/jcan.2000.9484 http://doi.org/10.1053/jcan.2000.9484 . Lidocaine for prevention of reperfusion ventricular fibrillation after release of aortic cross-clamping M Baravelli , P Cattaneo , A Rossi , 等 . Low-risk profile for malignant ventricular arrhythmias and sudden cardiac death after surgical ventricular reconstruction . Pacing Clin Electrophysiol , 2010 . 33 ( 9 ): 1054 - 1062 . 20487339 DOI: 10.1111/j.1540-8159.2010.02771.x http://doi.org/10.1111/j.1540-8159.2010.02771.x . Low-risk profile for malignant ventricular arrhythmias and sudden cardiac death after surgical ventricular reconstruction P Chevalier , Q Timour , E Morel , 等 . Chronic oral amiodarone but not dronedarone therapy increases ventricular defibrillation threshold during acute myocardial ischemia in a closed-chest animal model . J Cardiovasc Pharmacol , 2012 . 59 ( 6 ): 523 - 528 . 22330675 DOI: 10.1097/FJC.0b013e31824d89fe http://doi.org/10.1097/FJC.0b013e31824d89fe . Chronic oral amiodarone but not dronedarone therapy increases ventricular defibrillation threshold during acute myocardial ischemia in a closed-chest animal model PW Doherty , PR McLaughlin , M Billingham , 等 . Cardiac damage produced by direct current countershock applied to the heart . Am J Cardiol , 1979 . 43 ( 2 ): 225 - 232 . 760477 DOI: 10.1016/S0002-9149(79)80008-9 http://doi.org/10.1016/S0002-9149(79)80008-9 . Cardiac damage produced by direct current countershock applied to the heart P Dorian , D Cass , B Schwartz , 等 . Amiodarone as compared with lidocaine for shock-resistant ventricular fibrillation . N Engl J Med , 2002 . 347 ( 12 ): 955 DOI: 10.1056/NEJM200209193471227 http://doi.org/10.1056/NEJM200209193471227 . Amiodarone as compared with lidocaine for shock-resistant ventricular fibrillation AC Fiore , KS Naunheim , J Taub , 等 . Myocardial preservation using lidocaine blood cardioplegia . Ann Thorac Surg , 1990 . 50 ( 5 ): 771 - 775 . 2241341 DOI: 10.1016/0003-4975(90)90683-W http://doi.org/10.1016/0003-4975(90)90683-W . Myocardial preservation using lidocaine blood cardioplegia AD Furlan , A Malmivaara , R Chou , 等 . 2015 updated method guideline for systematic reviews in the Cochrane Back and Neck Group . Spine , 2015 . 40 ( 21 ): 1660 - 1673 . 26208232 DOI: 10.1097/BRS.0000000000001061 http://doi.org/10.1097/BRS.0000000000001061 . 2015 updated method guideline for systematic reviews in the Cochrane Back and Neck Group HJ Geissler , C Schlensak , M Sudkamp , 等 . Heart valve surgery today: indications, operative technique, and selected aspects of postoperative care in acquired valvular heart disease . Dtsch Arztebl Int , 2009 . 106 ( 13 ): 224 - 233 . 19471589 DOI: 10.3238/arztebl.2009.0224 http://doi.org/10.3238/arztebl.2009.0224 . Heart valve surgery today: indications, operative technique, and selected aspects of postoperative care in acquired valvular heart disease AA Ghavidel , S Nabavi , M Haghjoo , 等 . Amiodarone versus lidocaine for the prevention of reperfusion ventricular fibrillation: a randomized clinical trial . ARYA Atheroscler , 2013 . 1 343 - 349 . . Amiodarone versus lidocaine for the prevention of reperfusion ventricular fibrillation: a randomized clinical trial MJ Hippelainen , TT Tuppurainen , KT Huttunen . Reperfusion ventricular fibrillation and electric countershocks during coronary artery bypass operations-association with postoperative CK-MB release . J Thorac Cardiovasc Surg , 1994 . 28 ( 2 ): 73 - 78 . . Reperfusion ventricular fibrillation and electric countershocks during coronary artery bypass operations-association with postoperative CK-MB release J Kirlangitis , R Middaugh , R Knight , 等 . Comparison of bretylium and lidocaine in the prevention of ventricular fibrillation after aortic cross-clamp release in coronary artery bypass surgery . J Cardiothorac Anesth , 1990 . 4 ( 5 ): 582 - 587 . 2132137 DOI: 10.1016/0888-6296(90)90408-8 http://doi.org/10.1016/0888-6296(90)90408-8 . Comparison of bretylium and lidocaine in the prevention of ventricular fibrillation after aortic cross-clamp release in coronary artery bypass surgery S Komori , B Li , K Matsumura , 等 . Antiarrhythmic effect of magnesium sulfate against occlusion-induced arrhythmias and reperfusion-induced arrhythmias in anesthetized rats . Mol Cell Biochem , 1999 . 199 ( 1-2 ): 201 - 208 . 10544968 DOI: 10.1023/A:1006938010925 http://doi.org/10.1023/A:1006938010925 . Antiarrhythmic effect of magnesium sulfate against occlusion-induced arrhythmias and reperfusion-induced arrhythmias in anesthetized rats BP Leeuwenburgh , MI Versteegh , JJ Maas , 等 . Should amiodarone or lidocaine be given to patients who arrest after cardiac surgery and fail to cardiovert from ventricular fibrillation? . Interact Cardiovasc Thorac Surg , 2008 . 7 ( 6 ): 1148 - 1151 . 18796471 DOI: 10.1510/icvts.2008.188656 http://doi.org/10.1510/icvts.2008.188656 . Should amiodarone or lidocaine be given to patients who arrest after cardiac surgery and fail to cardiovert from ventricular fibrillation? JH Levine , A Massumi , MM Scheinman , 等 . Intravenous amiodarone for recurrent sustained hypotensive ventricular tachyarrhythmias intravenous amiodarone multicenter trial group . J Am Coll Cardiol , 1996 . 27 ( 1 ): 67 - 75 . 8522712 DOI: 10.1016/0735-1097(95)00427-0 http://doi.org/10.1016/0735-1097(95)00427-0 . Intravenous amiodarone for recurrent sustained hypotensive ventricular tachyarrhythmias intravenous amiodarone multicenter trial group XB Liu , JB Jiang , QJ Zhou , 等 . Evaluation of the safety and efficacy of transcatheter aortic valve implantation in patients with a severe stenotic bicuspid aortic valve in a Chinese population . J Zhejiang Univ-Sci B (Biomed & Biotechnol) , 2015 . 16 ( 3 ): 208 - 214 . 25743122 DOI: 10.1631/jzus.B1500017 http://doi.org/10.1631/jzus.B1500017 . Evaluation of the safety and efficacy of transcatheter aortic valve implantation in patients with a severe stenotic bicuspid aortic valve in a Chinese population SP Marathe , S Talwar . Surgery for transposition of great arteries: a historical perspective . Ann Pediatr Cardiol , 2015 . 8 ( 2 ): 122 - 128 . 26085763 DOI: 10.4103/0974-2069.157025 http://doi.org/10.4103/0974-2069.157025 . Surgery for transposition of great arteries: a historical perspective WJ Mauermann , JN Pulido , DW Barbara , 等 . Amiodarone versus lidocaine and placebo for the prevention of ventricular fibrillation after aortic crossclamping: a randomized, double-blind, placebo-controlled trial . J Thorac Cardiovasc Surg , 2012 . 144 ( 5 ): 1229 - 1234 . 22770549 DOI: 10.1016/j.jtcvs.2012.06.039 http://doi.org/10.1016/j.jtcvs.2012.06.039 . Amiodarone versus lidocaine and placebo for the prevention of ventricular fibrillation after aortic crossclamping: a randomized, double-blind, placebo-controlled trial CM Mery , LC Kane . The ACGME fellowship in congenital cardiac surgery: the graduates’ perspective . Semin Thorac Cardiovasc Surg Pediatr Card Surg Ann , 2017 . 20 70 - 76 . DOI: 10.1053/j.pcsu.2016.09.005 http://doi.org/10.1053/j.pcsu.2016.09.005 . The ACGME fellowship in congenital cardiac surgery: the graduates’ perspective D Moher , A Liberati , J Tetzlaff , 等 . Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement . Ann Intern Med , 2009 . 151 ( 4 ): 264 - 269 . 19622511 DOI: 10.7326/0003-4819-151-4-200908180-00135 http://doi.org/10.7326/0003-4819-151-4-200908180-00135 . Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement H Nayeem Ul , AM Dar , ML Wani , 等 . A comparative study on the effect of amiodarone and metaprolol for prevention of arrythmias after open heart surgery . Int Cardiovasc Res J , 2013 . 7 ( 1 ): 1 - 4 . 24757610 . A comparative study on the effect of amiodarone and metaprolol for prevention of arrythmias after open heart surgery OA Onk , B Erkut . Is the preoperative administration of amiodarone or metoprolol more effective in reducing atrial fibrillation: after coronary bypass surgery? . Medicine (Baltimore) , 2015 . 94 ( 41 ): e1576 26469896 DOI: 10.1097/MD.0000000000001576 http://doi.org/10.1097/MD.0000000000001576 . Is the preoperative administration of amiodarone or metoprolol more effective in reducing atrial fibrillation: after coronary bypass surgery? NA Papakonstantinou , NG Baikoussis , P Dedeilias , 等 . Cardiac surgery or interventional cardiology? Why not both? Let’s go hybrid . J Cardiol , 2017 . 69 ( 1 ): 46 - 56 . 27727088 DOI: 10.1016/j.jjcc.2016.09.007 http://doi.org/10.1016/j.jjcc.2016.09.007 . Cardiac surgery or interventional cardiology? Why not both? Let’s go hybrid KD Patil , HR Halperin , LB Becker . Cardiac arrest: resuscitation and reperfusion . Circ Res , 2015 . 116 ( 12 ): 2041 - 2049 . 26044255 DOI: 10.1161/CIRCRESAHA.116.304495 http://doi.org/10.1161/CIRCRESAHA.116.304495 . Cardiac arrest: resuscitation and reperfusion T Petrovic , F Adnet , C Lapandry . Successful resuscitation of ventricular fibrillation after low-dose amiodarone . Ann Emerg Med , 1998 . 32 ( 4 ): 518 - 519 . 9774944 DOI: 10.1016/S0196-0644(98)70191-X http://doi.org/10.1016/S0196-0644(98)70191-X . Successful resuscitation of ventricular fibrillation after low-dose amiodarone A Samantaray , A Chandra , S Panigrahi . Amiodarone for the prevention of reperfusion ventricular fibrillation . J Cardiothorac Vasc Anesth , 2010 . 24 ( 2 ): 239 - 243 . 19800815 DOI: 10.1053/j.jvca.2009.07.007 http://doi.org/10.1053/j.jvca.2009.07.007 . Amiodarone for the prevention of reperfusion ventricular fibrillation T Schluter , H Baum , A Plewan , 等 . Effects of implantable cardioverter defibrillator implantation and shock application on biochemical markers of myocardial damage . Clin Chem , 2001 . 47 ( 3 ): 459 - 463 . 11238297 . Effects of implantable cardioverter defibrillator implantation and shock application on biochemical markers of myocardial damage JG Shake , PJ Pronovost , GJ Whitman . Cardiac surgical ICU care: eliminating ‘preventable’ complications . J Card Surg , 2013 . 28 ( 4 ): 406 - 413 . 23718856 DOI: 10.1111/jocs.12124 http://doi.org/10.1111/jocs.12124 . Cardiac surgical ICU care: eliminating ‘preventable’ complications JC Somberg , SJ Bailin , CI Haffajee , 等 . Intravenous lidocaine versus intravenous amiodarone (in a new aqueous formulation) for incessant ventricular tachycardia . Am J Cardiol , 2002 . 90 ( 8 ): 853 - 859 . 12372573 DOI: 10.1016/S0002-9149(02)02707-8 http://doi.org/10.1016/S0002-9149(02)02707-8 . Intravenous lidocaine versus intravenous amiodarone (in a new aqueous formulation) for incessant ventricular tachycardia MTM Vaziri , R Jouibar , SHA Akhlagh , 等 . The effect of lidocaine and magnesium sulfate on prevention of ventricular fibrillation in coronary artery bypass grafting surgery . Iran Red Crescent Med J , 2010 . 12 ( 3 ): 298 - 301 . . The effect of lidocaine and magnesium sulfate on prevention of ventricular fibrillation in coronary artery bypass grafting surgery ED Verrier , MJ Mack . Viewpoint: transitions in cardiac surgery and interventional cardiology team mates or rivals? . Heart , 2015 . 101 ( 5 ): 346 - 348 . 25538137 DOI: 10.1136/heartjnl-2014-306132 http://doi.org/10.1136/heartjnl-2014-306132 . Viewpoint: transitions in cardiac surgery and interventional cardiology team mates or rivals? TL Wenger , S Lederman , CF Starmer , 等 . A method for quantitating antifibrillatory effects of drugs after coronary reperfusion in dogs: improved outcome with bretylium . Circulation , 1984 . 69 ( 1 ): 142 - 148 . 6689638 DOI: 10.1161/01.CIR.69.1.142 http://doi.org/10.1161/01.CIR.69.1.142 . A method for quantitating antifibrillatory effects of drugs after coronary reperfusion in dogs: improved outcome with bretylium L Wu , Q Jin , N Zhang , 等 . The effects of acute amiodarone on short-and long-duration ventricular defibrillation threshold in canines . J Cardiovasc Pharmacol , 2011 . 58 ( 4 ): 432 - 438 . 21709582 DOI: 10.1097/FJC.0b013e318228a50c http://doi.org/10.1097/FJC.0b013e318228a50c . The effects of acute amiodarone on short-and long-duration ventricular defibrillation threshold in canines MG Wyman , RM Wyman , DS Cannom , 等 . Prevention of primary ventricular fibrillation in acute myocardial infarction with prophylactic lidocaine . Am J Cardiol , 2004 . 94 ( 5 ): 545 - 551 . 15342281 DOI: 10.1016/j.amjcard.2004.05.014 http://doi.org/10.1016/j.amjcard.2004.05.014 . Prevention of primary ventricular fibrillation in acute myocardial infarction with prophylactic lidocaine H Yamaguchi , M Weil , W Tang , 等 . Myocardial dysfunction after electrical defibrillation . Resuscitation , 2002 . 54 ( 3 ): 289 - 296 . 12204463 DOI: 10.1016/S0300-9572(02)00149-1 http://doi.org/10.1016/S0300-9572(02)00149-1 . Myocardial dysfunction after electrical defibrillation ST Yeh , SE Aune , TA Wilgus , 等 . Hyperoxemic reperfusion after prolonged cardiac arrest in a rat cardiopulmonary bypass resuscitation model . Resuscitation , 2013 . 84 ( 1 ): 114 - 120 . 22982155 DOI: 10.1016/j.resuscitation.2012.08.323 http://doi.org/10.1016/j.resuscitation.2012.08.323 . Hyperoxemic reperfusion after prolonged cardiac arrest in a rat cardiopulmonary bypass resuscitation model M Yilmaz , U Aydin , ZI Arslan , 等 . The effect of lidocaine and amiodarone on prevention of ventricular fibrillation in patients undergoing coronary artery bypass grafting . Heart Surg Forum , 2014 . 17 ( 5 ): 245 - 249 . DOI: 10.1532/HSF98.2014402 http://doi.org/10.1532/HSF98.2014402 . The effect of lidocaine and amiodarone on prevention of ventricular fibrillation in patients undergoing coronary artery bypass grafting HT Zhan , ST Purcell , RL Bush . Preoperative optimization of the vascular surgery patient . Vasc Health Risk Manag , 2015 . 11 379 - 385 . 26170688 . Preoperative optimization of the vascular surgery patient F Zoerner , E Semenas . Resuscitation with amiodarone increases survival after hemorrhage and ventricular fibrillation in pigs . J Trauma Acute Care Surg , 2014 . 76 ( 6 ): 1402 - 1408 . 24854308 DOI: 10.1097/TA.0000000000000243 http://doi.org/10.1097/TA.0000000000000243 . Resuscitation with amiodarone increases survival after hemorrhage and ventricular fibrillation in pigs We thank Dr. Alan SEAL (New Zealand) for linguistic review of the manuscript. Compliance with ethics guidelines Yong ZHENG, Qiang GU, Hong-wu CHEN, Huai-ming PENG, Dong-yu JIA, Yu ZHOU, and Mei-xiang XIANG declare that they have no conflict of interest. This article does not contain any studies with human or animal subjects performed by any of the authors. 10.1631/jzus.B1700229 2017-04-24 2017-10-16 2022-01-05 Copyright © Zhejiang University and Springer-Verlag GmbH Germany 2017 2017 Journal of Zhejiang University-SCIENCE B (Biomedicine and Biotechnology) 12 18