- Email: [email protected]
Perioperative Statins Do Not Prevent Acute Kidney Injury After Cardiac Surgery: A Meta-analysis of Randomized Controlled Trials
Author’s Accepted Manuscript Perioperative Statins do not Prevent Acute Kidney Injury After Cardiac Surgery: A Meta-Analysis of Randomized Controlled Trials Bing-Cheng Zhao, Pu Shen, Ke-Xuan Liu www.elsevier.com/locate/buildenv
PII: DOI: Reference:
S1053-0770(17)30430-5 http://dx.doi.org/10.1053/j.jvca.2017.04.038 YJCAN4120
To appear in: Journal of Cardiothoracic and Vascular Anesthesia Cite this article as: Bing-Cheng Zhao, Pu Shen and Ke-Xuan Liu, Perioperative Statins do not Prevent Acute Kidney Injury After Cardiac Surgery: A MetaAnalysis of Randomized Controlled Trials, Journal of Cardiothoracic and Vascular Anesthesia, http://dx.doi.org/10.1053/j.jvca.2017.04.038 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Page 1
Perioperative statins do not prevent acute kidney injury after cardiac surgery: A metaanalysis of randomized controlled trials
Bing-Cheng Zhao, MD1, Pu Shen, MD2, Ke-Xuan Liu, MD, PhD1*
1
Department of Anesthesiology, Nanfang Hospital, Southern Medical University, 1838
Guangzhou Ave N, Guangzhou 510515, China 2
Department of Anesthesiology, the First Affiliated Hospital, Sun Yat-Sen University, 85
Zhongshan II Road, Guangzhou 510080, China
* Corresponding author: Prof. Ke-Xuan Liu, Department of Anesthesiology, Nanfang Hospital, Southern Medical University, 1838 Guangzhou Ave N, Guangzhou 510515, China; e-mail: [email protected]
Drs. Zhao and Shen contributed equally to this work.
Acknowledgments This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Page 2
Abstract OBJECTIVES: To evaluate whether perioperative statins reduce the risk of acute kidney injury (AKI) after cardiac surgery. DESIGN: Systematic review and meta-analysis of randomized trials. SETTING: Perioperative. PARTICIPANTS: Adult patients receiving cardiac surgery. INTERVENTIONS: PubMed, EMBASE and the Cochrane Library databases were searched for randomized trials. Random-effects meta-analyses were performed to compare the effects of statins versus placebo/control. Trial sequential analysis was conducted to confirm the results. MEASUREMENTS AND MAIN RESULTS: The primary outcome was incidence of postoperative AKI. Eight trials enrolling 3,204 patients were included. The statin arms and the control arms were comparable in incidence of postoperative AKI (risk ratio [RR] =1.02, 95% confidence interval [CI] =0.82-1.28), need for renal replacement therapy (RR=1.09, 95% CI=0.45-2.66), mechanical ventilation duration (mean difference [MD] =24.84 minutes, 95% CI=-55.53-105.20), ICU length of stay (MD=0.04 days, 95% CI=3.13-3.20), hospital length of stay (MD=-0.08 days, 95% CI=-0.31-0.15), and in-hospital mortality (RR=3.76, 95% CI=0.93-15.14). Trial sequential analysis confirmed that it is unlikely that perioperative statin therapy could achieve a 20 % or more relative risk reduction in AKI incidence. CONCLUSIONS: Among patients undergoing cardiac surgery, perioperative statin treatment did not reduce the risk of AKI. Statin therapy should not be initiated to prevent AKI following cardiac surgery.
Page 3
Keywords Acute kidney injury; Cardiac surgery; Statin; Meta-analysis
Introduction Acute kidney injury (AKI) is one of the most common complications after open-heart surgery, with up to 30% of patients developing this complication during their recovery [1]. Despite the great advancements in surgical techniques, anesthetic methods and critical care managements in the past decades, the incidence of postoperative AKI appears to have not changed significantly [2]. Because of demonstrated associations between the diagnosis of AKI and increased risk of postoperative morbidity (arrhythmias, respiratory failure, systemic infection, and myocardial infarction) and short- and long-term mortality [3-7], extensive efforts have been made to identify effective interventions to prevent postoperative AKI. Of the many potential prophylactic agents, Statins, 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors, would appear to be the ideal agent because they are relatively inexpensive, widely available, and have an excellent safety profile. Furthermore, several published meta-analyses have concluded that prophylactic administration of statins is effective at preventing atrial fibrillation, AKI and all-cause mortality after cardiac surgery [8-14].On the basis of this evidence, practice guidelines currently recommend that patients who undergo cardiac surgery without contraindication should receive statins preoperatively and continuously until surgery [15, 16]. However, when we reviewed the previously published meta-analyses, we noted that most included studies were retrospective and observational studies. Only 5 randomized controlled trials (RCT) on renal outcomes had been included [17-21], but most of them enrolled only a very small number of patients and did not reported postoperative AKI as a primary outcome. In the past few months, several large-scale RCTs have debated on the efficacy of statins in preventing AKI after cardiac surgery [22-24]. The information size
Page 4
regarding this issue has increased dramatically. Therefore, we conducted an updated metaanalysis with trial sequential analysis of RCTs to determine the prophylactic effects of statins against AKI following cardiac surgery. Material and methods The protocol of this study was developed according to the Cochrane Handbook for Systematic Reviews of Interventions and the PRISMA guidelines [25, 26]. Search strategy and selection criteria We searched PubMed, EMBASE and the Cochrane Library from inception to August 2016 without restriction on language (see details of the search strategy in Supplementary Table 1). The computer retrieval was supplemented by manual search of reference lists of identified studies and (systematic) reviews. Citations were first screened at the title and abstract level. Then full texts were retrieved if they reported potentially relevant studies. To be eligible, studies had to be randomized controlled trials with comparison of statin therapy (any duration and dosage) versus controls (placebo or no intervention) that started in perioperative period in adult (>18 years of age) patients, that were of cardiac surgery, and that reported incidence of postoperative kidney injury. Data extraction and quality assessment Two researchers reviewed the studies independently, with disagreements in eligibility, quality assessment and data extraction resolved by consensus. Using prepared extraction forms, data on study designs, patient characteristics, and outcomes were collected for each eligible RCT. The incidence of AKI in the postoperative period was chosen as the main outcome of interest because of its high incidence and its associations with other major morbidities after cardiac surgery [7]. Secondary outcomes assessed included the need for renal replacement therapy, mechanical ventilation duration, length of intensive care unit (ICU) and hospital stays, and in-hospital mortality. Trials included in the meta-analysis were
Page 5
evaluated for methodological quality using the Cochrane Collaboration’s tool [25], which consists of seven domains: sequence generation; allocation concealment; blinding of participants and personnel; blinding of outcome assessors; incomplete outcome data; selective outcome reporting; and other bias. We categorized trials with low overall risk of bias as those with a low risk of bias in all domains. Other trials were categorized as unclear or at high risk of bias. Statistical analysis Whenever available, we used results from intention-to-treat analysis. The differences in dichotomous outcomes were reported as relative risk (RR) with 95% confidence interval (CI) using a random-effects model. The differences in continuous outcomes were reported as mean difference (MD) with 95% CI also using a random-effects model. Heterogeneity across studies was assessed using the chi-square test and was quantified by I2 statistic. Heterogeneity was perceived as low (I2 = 25%-50%), moderate (I2 = 50%-75%), or high (I2 >75%). Publication bias was assessed by visual inspection of the funnel plots. For the primary outcome, we performed several sensitivity analyses, including fitting a fixed-effect model, excluding studies with high and unclear risk of bias, including only studies that used atorvastatin, including only patients naive to statin therapy, including only studies with AKI as the primary outcome, and including only studies that used RIFLE (risk, injury, failure, loss, and end-stage kidney) or AKIN (AKI Network) criteria to define AKI. Additionally, sensitivity analysis by omitting one trial at a time and pooling the others was performed to detect the influence of a single trial on the overall estimate. Review Manager 5.3.3 (Cochrane Collaboration, Nordic Cochrane Centre, Denmark) was used for the above calculations. Cumulative meta-analyses may result in type I errors due to repeated significance testing when updated with new trials. We examined the reliability and conclusiveness of our results using trial sequential analysis. It is similar to interim analyses in a single trial, where
Page 6
monitoring boundaries are used to decide whether the trial could be terminated early when a P value is sufficiently small to show the anticipated effect or sufficiently large to show potential futility. Because no reason exists why the standards for a meta-analysis could be less rigorous than those for a single trial, similar trial sequential monitoring boundaries can be applied to meta-analysis as trial sequential analysis. It can provide information as to whether to continue evaluating for evidence when the boundary is not crossed or whether sufficient evidence is reached for anticipated effect or for futility when the boundary is crossed [27]. In this study, the assumptions included two-sided testing, type I error of 5% and power of 80%. Diversity-adjusted information size was calculated based on a control event rates (the absolute AKI incidence in the control arms of included trials) and relative risk reduction of 20%. The 20% relative risk reduction was chosen because it is an effect size that is both clinically meaningful and realistic in cardiovascular trials. Boundaries for concluding superiority or inferiority or futility were calculated with the O'Brien–Fleming α-spending function. This analysis was carried out using a specific software (User Manual for TSA; Copenhagen Trial Unit 2011, Denmark). Results Search results and study characteristics Fig. 1 is a flow diagram of the study and summarizes the process of RCT selection. Eight eligible RCTs were identified, which enrolled 3,204 patients randomly assigned to receive statin therapy or placebo/no intervention. The earliest study was published in 1999 and the 3 most recent studies were published in 2016. Study sample sizes varied from 40 to 1,922. Main characteristics of included trials on study design, surgery type, patient characteristic, statin therapy (varied widely in terms of drug, dose, timing of administration, and duration) and AKI definition is listed in Table 1.
Page 7
Assessment of risk of bias Risk of bias assessment found 4 studies with low risk of bias. One study was considered to have high risk of bias because the authors did not clearly address reporting of incomplete outcome data. Other 3 studies had unclear risk of bias because random sequence generation and/or allocation concealment were not clearly described (Fig. 2). Primary outcome: incidence of AKI Postoperative AKI was reported in all included studies, although different definitions were applied by the investigators of respective trials, including the AKIN criterion [22-24], RIFLE criterion [21], and others [17-20]. Postoperative AKI occurred in 341 of the 1,603 patients receiving statin therapy and in 301 of the 1,601 patients in the control arms. Analyzed with a random-effects model, the pooled estimate of the eight trials suggested that perioperative statin therapy was not associated with a significant reduction in the incidence of postoperative AKI (RR=1.02, 95% CI=0.82-1.28, P=0.85; heterogeneity I2=26%; Fig. 2). Visual inspection of the funnel plot showed some asymmetry (Supplementary Fig. 1), but we did not conduct statistical tests for the asymmetry due to the small number of included trials [25]. Various sensitivity analyses were performed for the primary outcome. Meta-analysis using a fixed-effects model did not show significantly different AKI incidence in statin therapy arms and in control arms. Besides, meta-analyses were repeated after excluding studies with unclear and high risk of bias, excluding studies that used statins other than atorvastatin, excluding studies that did not report AKI as a primary outcome, excluding studies that did not use RIFLE or AKIN criteria for AKI diagnosis, and excluding patients that were chronic statin users. None of these found a significant effect of statin therapy on
Page 8
AKI incidence (Supplementary Table 2). Furthermore, sensitivity analysis by omitting one trial at a time showed that none of the included trials markedly changed the overall result. We conducted trial sequential analysis of statin therapy versus control on postoperative AKI incidence (Fig. 3). Based on a control event rate of 18.8% (301 events in 1,601 patients) and an a priori anticipated intervention effect of a relative risk reduction of 20 %, trial sequential analysis estimated that the optimal sample size to draw a conclusion was 8,746. In the present meta-analysis, the number of patients enrolled in all trials was 3,204 and the overall estimate was not statistically significant. The cumulative Z curve breaks through the boundary for futility (nonsuperiority). Thus, an intervention effect of 20 % relative risk reduction or more on AKI incidence is unlikely. Secondary outcomes The results of analyses on secondary outcomes are summarized in Supplementary Table 3. These outcomes were only reported in some of the included studies. Compared with control, perioperative statin therapy did not significantly alter the clinical outcomes of the need for renal replacement therapy (RR=1.09, 95% CI=0.45-2.66; number of studies n=3), mechanical ventilation duration (MD=24.84 min, 95% CI=-55.53-105.20; n=2), ICU length of stay (MD=0.04 days, 95% CI=-3.13-3.20, n=7), hospital length of stay (MD=-0.08 days, 95% CI=-0.31-0.15; n=6), or in-hospital mortality (RR=3.76, 95% CI=0.93-15.14; n=7). Discussion In this meta-analysis consisting of eight RCTs that enrolled a total of 3,204 patients undergoing open-heart surgery, we found that perioperative statin therapy did not significantly affect the incidence of AKI after surgery. Trial sequential analysis suggested that sufficient evidence had accrued to reject the possibility that perioperative statin therapy could achieve a 20 % or more relative risk reduction in AKI incidence. Statin had also failed
Page 9
in improving other postoperative outcomes including the need for renal replacement therapy, mechanical ventilation duration, length of stays in ICU and hospital, and in-hospital mortality. Previously, a Cochrane review had summarized RCTs that assessed the impact of preoperative statin use for surgery requiring cardiac bypass on the incidence of postoperative AKI [28]. It showed that the risk of AKI was nominally lower in the statin group than the control group, but the benefit did not reach statistical significance (RR=0.76, 95% CI= 0.461.28). However, this study included only five small RCTs that enrolled 467 patients in total. Several other relevant review articles published in recent years included observational studies in their analyses [9-11, 14]. Many, but not all, of them showed that perioperative statin therapy was associated with a significant risk reduction in postoperative AKI. However, the results may have been biased because most patients included in these studies were from observational studies while data from RCTs only contributed a negligible part. Observational studies frequently reach distorted conclusions because they are influenced by confounding. Conclusive evidence on this clinical issue should be addressed by well-conducted large-scale randomized trials. Fortunately, with the recent publication of several RCTs, the information size regarding this issue has dramatically increased, making an updated meta-analysis warranted. The result of the present meta-analysis is consistent with that of the Cochrane review that perioperative statin therapy brings no benefit in reducing AKI following openheat surgery. Another meta-analysis of RCTs that included recently published trials and reported the outcome of AKI was noticed [29]. In that study, meta-analysis based primarily on three trials with low risk of bias reached a conclusion that statin therapy was associated with an increase in incidence of AKI after cardiac surgery. The result was driven by a highly influential trial, which got 70.2% of the weight [24]. However, this trial did not study postoperative AKI as a primary outcome, and the investigators did not control for multiple comparisons. Thus, the
Page 10
probability of incorrect rejections of the null hypotheses (type I errors) was high [30]. On the contrary, our meta-analysis including all available trials showed statin therapy did not change the risk of AKI. This is supported by additional analyses, including (1) the trial sequential analysis showed enough evidence had accumulated that statin is unlikely to influence the incidence of AKI, (2) various sensitivity analyses of trials/patients with different characteristics showed non-significant results, and (3) other outcomes related with AKI, such as need for renal replacement therapy, length of stay and mortality, were unchanged by the statin therapy. Statins are commonly used lipid-lowering drugs, but their non-lipid-lowering activities, such as anti-inflammatory, antioxidative, and endothelial protecting effects, made them a potential agent for protecting renal function [31, 32]. There is growing evidence supporting the effectiveness of statins in reducing contrast-induced AKI in patients undergoing diagnostic or interventional procedures with contrast media [33, 34]. In cardiac surgery, however, statins’ renal protecting effect has only been shown in some observational studies [35, 36]. None of the three RCTs that studied AKI as a primary outcome demonstrated a significantly reduced postoperative AKI incidence in the statin arms [22-24], and our present meta-analysis of eight RCTs has confirmed this conclusion. The reason why statins can prevent contrast-induced AKI but not cardiac surgery-associated AKI is unknown. Similar observations have also been made in trials of N-acetylcysteine and sodium bicarbonate. Although some uncertainty remains [37, 38], a number of randomized trials as well as metaanalyses support the effectiveness of N-acetylcysteine and sodium bicarbonate therapy for the prevention of contrast-induced AKI [39, 40, 41, 42]. However, both of them have failed to prevent AKI after open-heart surgery [43, 44]. It is possible that the discrepancy in prophylactic efficacies is related to the different degree of inflammatory reaction in the two clinical settings. These medications alone might not be sufficient to protect the kidney against
Page 11
marked systemic inflammation and oxidative stress induced in heart surgery. Another possible reason is that some type of statins, if not all, might have adverse effects on renal function. For example, rosuvastatin treatment has been associated with new proteinuria onset [45].These adverse effects might have offset the potential renoprotective effects of statins through their anti-inflammatory and antioxidant activities. In addition, some authors suggested that statin-naive patients, who tend to have less cardiovascular morbidity, might get limited benefit from statins [23]. Our results should be interpreted in light of some limitations. First of all, the definitions of AKI were different across included studies. Some used serum creatinine cut-offs to dichotomize patients into AKI or no AKI. However, when we excluded studies that did not use the currently recommended criteria (RIFLE or AKIN) in a sensitivity analysis, the results remained unchanged. Second, the amount of evidence concerning the secondary outcomes such as the need for renal replacement is small and firm conclusions could not be reached. Third, the statin therapy used in individual studies varied widely in terms of drug, dose, timing of administration, and duration. We could not perform subgroup analysis on these factors due to the limited number of studies. Furthermore, the safety and efficacy of prophylactic statins might be different in specific patient subgroups, such as long-term statin users, those with chronic kidney disease or diabetes, and patients receiving on-pump or offpump surgeries. In this study-level meta-analysis, we could not examine the effects of these variables. The renal effects of perioperative statin therapy should be further studied for specific statin regimens and patient groups. We searched the ClincalTrials.gov website and noticed some ongoing trials on statin treatment in cardiac surgery. For example, the StaRT-CABG trial (NCT01715714) plans to enroll 2,630 patients to study the effects of high doses atorvastatin before coronary artery bypass grafting in chronic statin users. The results of these
Page 12
trials will answer part of this question. Besides, the subgroup analyses of trials are usually underpowered to identify subgroup effects. A meta-analysis of individual patient data from existing clinical trials would provide greater statistical power [46]. Such analysis may shed more light on the true effects of statin treatment in specific patient and care-delivery subgroups. In conclusion, in this updated meta-analysis with trial sequential analysis of RCTs of patients undergoing cardiac surgery, perioperative statin treatment did not reduce the risk of AKI. The current evidence does not support the use of statin therapy to prevent AKI following cardiac surgery.
Page 13
References [1] Rosner MH, Okusa MD. Acute kidney injury associated with cardiac surgery. Clin J Am Soc Nephrol 2006;1:19-32. [2] Siddiqui NF, Coca SG, Devereaux PJ, Jain AK, Li L, Luo J, et al. Secular trends in acute dialysis after elective major surgery-1995 to 2009. CMAJ 2012;184:1237-45. [3] Birnie K, Verheyden V, Pagano D, Bhabra M, Tilling K, Sterne JA, et al. Predictive models for kidney disease: improving global outcomes (KDIGO) defined acute kidney injury in UK cardiac surgery. Crit Care 2014;18:606. [4] Thakar CV, Yared JP, Worley S, Cotman K, Paganini EP. Renal dysfunction and serious infections after open-heart surgery. Kidney Int 2003;64:239-46. [5] Hoste EA, Bagshaw SM, Bellomo R, Cely CM, Colman R, Cruz DN, et al. Epidemiology of acute kidney injury in critically ill patients: the multinational AKI-EPI study. Intensive Care Med 2015;41:1411-23. [6] Lafrance JP, Miller DR. Acute kidney injury associates with increased long-term mortality. J Am Soc Nephrol 2010;21:345-52. [7] Pickering JW, James MT, Palmer SC. Acute kidney injury and prognosis after cardiopulmonary bypass: a meta-analysis of cohort studies. Am J Kidney Dis 2015;65:283-93. [8] Liakopoulos OJ, Choi YH, Haldenwang PL, Strauch J, Wittwer T, Dorge H, et al. Impact of preoperative statin therapy on adverse postoperative outcomes in patients undergoing cardiac surgery: a meta-analysis of over 30,000 patients. Eur Heart J 2008;29:1548-59. [9] Wang J, Gu C, Gao M, Yu W, Yu Y. Preoperative Statin Therapy and Renal Outcomes After Cardiac Surgery: A Meta-analysis and Meta-regression of 59,771 Patients. Can J Cardiol 2015;31:1051-60.
Page 14
[10] Pan SY, Wu VC, Huang TM, Chou HC, Ko WJ, Wu KD, et al. Effect of preoperative statin therapy on postoperative acute kidney injury in patients undergoing major surgery: systemic review and meta-analysis. Nephrology (Carlton) 2014;19:750-63. [11] Kuhn EW, Liakopoulos OJ, Stange S, Deppe AC, Slottosch I, Choi YH, et al. Preoperative statin therapy in cardiac surgery: a meta-analysis of 90,000 patients. Eur J Cardiothorac Surg 2014;45:17-26; discussion [12] Singh I, Rajagopalan S, Srinivasan A, Achuthan S, Dhamija P, Hota D, et al. Preoperative statin therapy is associated with lower requirement of renal replacement therapy in patients undergoing cardiac surgery: a meta-analysis of observational studies. Interact Cardiovasc Thorac Surg 2013;17:345-52. [13] Kuhn EW, Slottosch I, Wahlers T, Liakopoulos OJ. Preoperative statin therapy for patients undergoing cardiac surgery. Cochrane Database Syst Rev 2015:CD008493. [14] Li M, Zou H, Xu G. The prevention of statins against AKI and mortality following cardiac surgery: A meta-analysis. Int J Cardiol 2016;222:260-6. [15] Hillis LD, Smith PK, Anderson JL, Bittl JA, Bridges CR, Byrne JG, et al. 2011 ACCF/AHA Guideline for Coronary Artery Bypass Graft Surgery. A report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. Developed in collaboration with the American Association for Thoracic Surgery, Society of Cardiovascular Anesthesiologists, and Society of Thoracic Surgeons. J Am Coll Cardiol 2011;58:e123-210. [16] Authors/Task Force members, Windecker S, Kolh P, Alfonso F, Collet JP, Cremer J, et al. 2014 ESC/EACTS Guidelines on myocardial revascularization: The Task Force on Myocardial Revascularization of the European Society of Cardiology (ESC) and the European Association for Cardio-Thoracic Surgery (EACTS)Developed with the special
Page 15
contribution of the European Association of Percutaneous Cardiovascular Interventions (EAPCI). Eur Heart J 2014;35:2541-619. [17] Christenson JT. Preoperative lipid-control with simvastatin reduces the risk of postoperative thrombocytosis and thrombotic complications following CABG. Eur J Cardiothorac Surg 1999;15:394-9; discussion 9-400. [18] Spadaccio C, Pollari F, Casacalenda A, Alfano G, Genovese J, Covino E, et al. Atorvastatin increases the number of endothelial progenitor cells after cardiac surgery: a randomized control study. J Cardiovasc Pharmacol 2010;55:30-8. [19] Chello M, Patti G, Candura D, Mastrobuoni S, Di Sciascio G, Agro F, et al. Effects of atorvastatin on systemic inflammatory response after coronary bypass surgery. Crit Care Med 2006;34:660-7. [20] Mannacio VA, Iorio D, De Amicis V, Di Lello F, Musumeci F. Effect of rosuvastatin pretreatment on myocardial damage after coronary surgery: a randomized trial. J Thorac Cardiovasc Surg 2008;136:1541-8. [21] Prowle JR, Calzavacca P, Licari E, Ligabo EV, Echeverri JE, Haase M, et al. Pilot double-blind, randomized controlled trial of short-term atorvastatin for prevention of acute kidney injury after cardiac surgery. Nephrology (Carlton) 2012;17:215-24. [22] Billings FTt, Hendricks PA, Schildcrout JS, Shi Y, Petracek MR, Byrne JG, et al. Highdose perioperative atorvastatin and acute kidney injury following cardiac surgery: A randomized clinical trial. JAMA 2016;315:877-88. [23] Park JH, Shim JK, Song JW, Soh S, Kwak YL. Effect of atorvastatin on the incidence of acute kidney injury following valvular heart surgery: a randomized, placebo-controlled trial. Intensive Care Med 2016;42:1398-407. [24] Zheng Z, Jayaram R, Jiang L, Emberson J, Zhao Y, Li Q, et al. Perioperative rosuvastatin in cardiac surgery. N Engl J Med 2016;374:1744-53.
Page 16
[25] Higgins J, Green S. Cochrane handbook for systematic reviews of interventions, Version 5.1.0 (Updated March 2011), http://www.cochrane-handbook.org (accessed September 12, 2016) [26] Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 2009;6:e1000097. . [27] Wetterslev J, Thorlund K, Brok J, Gluud C. Trial sequential analysis may establish when firm evidence is reached in cumulative meta-analysis. J Clin Epidemiol 2008;61:64-75. [28] Lewicki M, Ng I, Schneider AG. HMG CoA reductase inhibitors (statins) for preventing acute kidney injury after surgical procedures requiring cardiac bypass. Cochrane Database Syst Rev 2015:CD010480. [29] Putzu A, Capelli B, Belletti A, et al. Perioperative statin therapy in cardiac surgery: a meta-analysis of randomized controlled trials. Crit Care 2016; 20:395. [30] Aickin M, Gensler H. Adjusting for multiple testing when reporting research results: the Bonferroni vs Holm methods. Am J Public Health 1996;86:726-8. [31] Epstein M, Campese VM. Pleiotropic effects of 3-hydroxy-3-methylglutaryl coenzyme a reductase inhibitors on renal function. Am J Kidney Dis 2005;45:2-14. [32] Campese VM, Nadim MK, Epstein M. Are 3-hydroxy-3-methylglutaryl-CoA reductase inhibitors renoprotective? J Am Soc Nephrol 2005;16 Suppl 1:S11-7. [33] Jo SH, Koo BK, Park JS, Kang HJ, Cho YS, Kim YJ, et al. Prevention of radiocontrast medium-induced nephropathy using short-term high-dose simvastatin in patients with renal insufficiency undergoing coronary angiography (PROMISS) trial - a randomized controlled study. Am Heart J 2008;155. [34] Su X, Xie X, Liu L, Lv J, Song F, Perkovic V, et al. Comparative effectiveness of 12 treatment strategies for preventing contrast-induced acute kidney injury: A systematic review and Bayesian network meta-analysis. Am J Kidney Dis 2017;69:69-77.
Page 17
[35] Layton JB, Kshirsagar AV, Simpson RJ, Pate V, Funk MJ, Sturmer T, et al. Effect of statin use on acute kidney injury risk following coronary artery bypass grafting. Am J Cardiol 2013;111:823-8. [36] Layton JB, Hansen MK, Jakobsen CJ, Kshirsagar AV, Andreasen JJ, Hjortdal VE, et al. Statin initiation and acute kidney injury following elective cardiovascular surgery: a population cohort study in Denmark. Eur J Cardio-Thoracic Surg 2016;49:995-1000. [37] ACT Investigators. Acetylcysteine for prevention of renal outcomes in patients undergoing coronary and peripheral vascular angiography: main results from the randomized Acetylcysteine for Contrast-induced nephropathy Trial (ACT). Circulation. 2011;124:1250-9. [38] Lee SW, Kim WJ, Kim YH, Park SW, Park DW, Yun SC, et al. Preventive strategies of renal insufficiency in patients with diabetes undergoing intervention or arteriography (the PREVENT trial). Am J Cardiol 2011;107:1447-52. [39] Marenzi G, Assanelli E, Marana I, Lauri G, Campodonico J, Grazi M, et al. Nacetylcysteine and contrast-induced nephropathy in primary angioplasty. N Engl J Med 2006;354:2773-82. [40] Merten GJ, Burgess WP, Gray LV, Holleman JH, Roush TS, Kowalchuk GJ, et al. Prevention of contrast-induced nephropathy with sodium bicarbonate: a randomized controlled trial. JAMA 2004;291:2328-34. [41] Xu R, Tao A, Bai Y, Deng Y, Chen G. Effectiveness of N-acetylcysteine for the prevention of contrast-induced nephropathy: A systematic review and meta-analysis of randomized controlled trials. J Am Heart Assoc 2016;5: e003968. [42] Jang JS, Jin HY, Seo JS, Yang TH, Kim DK, Kim TH, et al. Sodium bicarbonate therapy for the prevention of contrast-induced acute kidney injury – a systematic review and metaanalysis –. Circ J 2012;76:2255-65.
Page 18
[43] Adabag AS, Ishani A, Koneswaran S, Johnson DJ, Kelly RF, Ward HB, et al. Utility of N-acetylcysteine to prevent acute kidney injury after cardiac surgery: a randomized controlled trial. Am Heart J 2008;155:1143-9. [44] Haase M, Haase-Fielitz A, Plass M, Kuppe H, Hetzer R, Hannon C, et al. Prophylactic perioperative sodium bicarbonate to prevent acute kidney injury following open heart surgery: a multicenter double-blinded randomized controlled trial. PLoS Med 2013;10:e1001426. [45] Savarese G, Musella F, Volpe M, Paneni F, Perrone-Filardi P. Effects of atorvastatin and rosuvastatin on renal function: a meta-analysis. Int J Cardiol 2013;167:2482-9. [46] Reade MC, Delaney A, Bailey MJ, Harrison DA, Yealy DM, Jones PG, et al. Prospective meta-analysis using individual patient data in intensive care medicine. Intensive Care Med 2010;36:11-21.
Fig. 1. PRISMA flow diagram of the study. Fig. 2. Meta-analysis of perioperative statins versus control for acute kidney injury prevention: forest plot of the risk ratio of acute kidney injury in patients treated with statins compared with patients treated with control; risk of bias assessment of included studies. Fig. 3. Trial sequential analysis of included studies with a control event rate of 18.8%, type I error of 5%, power of 80%, and anticipated relative risk reduction of 20%.
Page 19
Table 1. Basic characteristics of randomized trials included in meta-analysis Author, Study Surgery CPB Age Sex Chronic Statins type, dosage, year, use (yrs) (M, %) statins design type timing reference (%) use (%) Christenson OL, R CABG 1999 [17]
100 63
80.5
NR
simvastatin 20 mg/d for 4w before surgery
Chello 2006 [19] Mannacio 2008 [20]
DB, CABG PC, R DB, CABG PC, R
100 65
77.5
0
100 60
72.5
NR
atorvastatin 20 mg/d for 3w before surgery rosuvastatin 20 mg/d for 7d before surgery
Spadaccio 2010 [18] Prowle 2012 [21]
DB, CABG 100 65 PC, R DB, CABG, 100 68 PC, R valvular
54.0
0
70.0
70
Billings DB, CABG, 70.8 67 2016 [22] PC, R valvular
69.4
67.6
Park 2016 DB, valvular 100 58 [23] PC, R
49.5
0
Zheng DB, CABG, 53.2 59 2016 [24] PC, R valvular
79.2
34.0
atorvastatin 20 mg/d for 3w before surgery atorvastatin 40 mg the morning of surgery and on postoperative days 1-3 atorvastatin 80 mg the day before surgery, 40 mg the morning of surgery, and 40 mg/d after surgery until discharge atorvastatin 80 mg the evening before surgery, 40 mg the morning of surgery, and 40 mg/d on postoperative days 0-2 rosuvastatin 20 mg/d for up to 8 days before surgery and for 5 days thereafter
AKI definition serum urea ≥ 9 mmol/L and SCr ≥ 125 µmol/L NR increase in SCr > 221 µmol/L NR RIFLE category R or greater AKIN criteria stage 1 or greater
AKIN criteria stage 1 or greater
AKIN criteria stage 1 or greater
AKI = acute kidney injury, AKIN = Acute Kidney Injury Network, CABG = coronary artery bypass grafting, CPB = cardiopulmonary bypass, DB = double blind, M = male, NR = not reported, OL = open label, PC = placebo controlled, R = randomized, RIFLE = Risk of renal injury/Injury to the kidney/Failure of kidney function/Loss of kidney function/End stage disease, SCr = serum creatinine.
Page 20
Page 21
PII: DOI: Reference:
S1053-0770(17)30430-5 http://dx.doi.org/10.1053/j.jvca.2017.04.038 YJCAN4120
To appear in: Journal of Cardiothoracic and Vascular Anesthesia Cite this article as: Bing-Cheng Zhao, Pu Shen and Ke-Xuan Liu, Perioperative Statins do not Prevent Acute Kidney Injury After Cardiac Surgery: A MetaAnalysis of Randomized Controlled Trials, Journal of Cardiothoracic and Vascular Anesthesia, http://dx.doi.org/10.1053/j.jvca.2017.04.038 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Page 1
Perioperative statins do not prevent acute kidney injury after cardiac surgery: A metaanalysis of randomized controlled trials
Bing-Cheng Zhao, MD1, Pu Shen, MD2, Ke-Xuan Liu, MD, PhD1*
1
Department of Anesthesiology, Nanfang Hospital, Southern Medical University, 1838
Guangzhou Ave N, Guangzhou 510515, China 2
Department of Anesthesiology, the First Affiliated Hospital, Sun Yat-Sen University, 85
Zhongshan II Road, Guangzhou 510080, China
* Corresponding author: Prof. Ke-Xuan Liu, Department of Anesthesiology, Nanfang Hospital, Southern Medical University, 1838 Guangzhou Ave N, Guangzhou 510515, China; e-mail: [email protected]
Drs. Zhao and Shen contributed equally to this work.
Acknowledgments This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Page 2
Abstract OBJECTIVES: To evaluate whether perioperative statins reduce the risk of acute kidney injury (AKI) after cardiac surgery. DESIGN: Systematic review and meta-analysis of randomized trials. SETTING: Perioperative. PARTICIPANTS: Adult patients receiving cardiac surgery. INTERVENTIONS: PubMed, EMBASE and the Cochrane Library databases were searched for randomized trials. Random-effects meta-analyses were performed to compare the effects of statins versus placebo/control. Trial sequential analysis was conducted to confirm the results. MEASUREMENTS AND MAIN RESULTS: The primary outcome was incidence of postoperative AKI. Eight trials enrolling 3,204 patients were included. The statin arms and the control arms were comparable in incidence of postoperative AKI (risk ratio [RR] =1.02, 95% confidence interval [CI] =0.82-1.28), need for renal replacement therapy (RR=1.09, 95% CI=0.45-2.66), mechanical ventilation duration (mean difference [MD] =24.84 minutes, 95% CI=-55.53-105.20), ICU length of stay (MD=0.04 days, 95% CI=3.13-3.20), hospital length of stay (MD=-0.08 days, 95% CI=-0.31-0.15), and in-hospital mortality (RR=3.76, 95% CI=0.93-15.14). Trial sequential analysis confirmed that it is unlikely that perioperative statin therapy could achieve a 20 % or more relative risk reduction in AKI incidence. CONCLUSIONS: Among patients undergoing cardiac surgery, perioperative statin treatment did not reduce the risk of AKI. Statin therapy should not be initiated to prevent AKI following cardiac surgery.
Page 3
Keywords Acute kidney injury; Cardiac surgery; Statin; Meta-analysis
Introduction Acute kidney injury (AKI) is one of the most common complications after open-heart surgery, with up to 30% of patients developing this complication during their recovery [1]. Despite the great advancements in surgical techniques, anesthetic methods and critical care managements in the past decades, the incidence of postoperative AKI appears to have not changed significantly [2]. Because of demonstrated associations between the diagnosis of AKI and increased risk of postoperative morbidity (arrhythmias, respiratory failure, systemic infection, and myocardial infarction) and short- and long-term mortality [3-7], extensive efforts have been made to identify effective interventions to prevent postoperative AKI. Of the many potential prophylactic agents, Statins, 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors, would appear to be the ideal agent because they are relatively inexpensive, widely available, and have an excellent safety profile. Furthermore, several published meta-analyses have concluded that prophylactic administration of statins is effective at preventing atrial fibrillation, AKI and all-cause mortality after cardiac surgery [8-14].On the basis of this evidence, practice guidelines currently recommend that patients who undergo cardiac surgery without contraindication should receive statins preoperatively and continuously until surgery [15, 16]. However, when we reviewed the previously published meta-analyses, we noted that most included studies were retrospective and observational studies. Only 5 randomized controlled trials (RCT) on renal outcomes had been included [17-21], but most of them enrolled only a very small number of patients and did not reported postoperative AKI as a primary outcome. In the past few months, several large-scale RCTs have debated on the efficacy of statins in preventing AKI after cardiac surgery [22-24]. The information size
Page 4
regarding this issue has increased dramatically. Therefore, we conducted an updated metaanalysis with trial sequential analysis of RCTs to determine the prophylactic effects of statins against AKI following cardiac surgery. Material and methods The protocol of this study was developed according to the Cochrane Handbook for Systematic Reviews of Interventions and the PRISMA guidelines [25, 26]. Search strategy and selection criteria We searched PubMed, EMBASE and the Cochrane Library from inception to August 2016 without restriction on language (see details of the search strategy in Supplementary Table 1). The computer retrieval was supplemented by manual search of reference lists of identified studies and (systematic) reviews. Citations were first screened at the title and abstract level. Then full texts were retrieved if they reported potentially relevant studies. To be eligible, studies had to be randomized controlled trials with comparison of statin therapy (any duration and dosage) versus controls (placebo or no intervention) that started in perioperative period in adult (>18 years of age) patients, that were of cardiac surgery, and that reported incidence of postoperative kidney injury. Data extraction and quality assessment Two researchers reviewed the studies independently, with disagreements in eligibility, quality assessment and data extraction resolved by consensus. Using prepared extraction forms, data on study designs, patient characteristics, and outcomes were collected for each eligible RCT. The incidence of AKI in the postoperative period was chosen as the main outcome of interest because of its high incidence and its associations with other major morbidities after cardiac surgery [7]. Secondary outcomes assessed included the need for renal replacement therapy, mechanical ventilation duration, length of intensive care unit (ICU) and hospital stays, and in-hospital mortality. Trials included in the meta-analysis were
Page 5
evaluated for methodological quality using the Cochrane Collaboration’s tool [25], which consists of seven domains: sequence generation; allocation concealment; blinding of participants and personnel; blinding of outcome assessors; incomplete outcome data; selective outcome reporting; and other bias. We categorized trials with low overall risk of bias as those with a low risk of bias in all domains. Other trials were categorized as unclear or at high risk of bias. Statistical analysis Whenever available, we used results from intention-to-treat analysis. The differences in dichotomous outcomes were reported as relative risk (RR) with 95% confidence interval (CI) using a random-effects model. The differences in continuous outcomes were reported as mean difference (MD) with 95% CI also using a random-effects model. Heterogeneity across studies was assessed using the chi-square test and was quantified by I2 statistic. Heterogeneity was perceived as low (I2 = 25%-50%), moderate (I2 = 50%-75%), or high (I2 >75%). Publication bias was assessed by visual inspection of the funnel plots. For the primary outcome, we performed several sensitivity analyses, including fitting a fixed-effect model, excluding studies with high and unclear risk of bias, including only studies that used atorvastatin, including only patients naive to statin therapy, including only studies with AKI as the primary outcome, and including only studies that used RIFLE (risk, injury, failure, loss, and end-stage kidney) or AKIN (AKI Network) criteria to define AKI. Additionally, sensitivity analysis by omitting one trial at a time and pooling the others was performed to detect the influence of a single trial on the overall estimate. Review Manager 5.3.3 (Cochrane Collaboration, Nordic Cochrane Centre, Denmark) was used for the above calculations. Cumulative meta-analyses may result in type I errors due to repeated significance testing when updated with new trials. We examined the reliability and conclusiveness of our results using trial sequential analysis. It is similar to interim analyses in a single trial, where
Page 6
monitoring boundaries are used to decide whether the trial could be terminated early when a P value is sufficiently small to show the anticipated effect or sufficiently large to show potential futility. Because no reason exists why the standards for a meta-analysis could be less rigorous than those for a single trial, similar trial sequential monitoring boundaries can be applied to meta-analysis as trial sequential analysis. It can provide information as to whether to continue evaluating for evidence when the boundary is not crossed or whether sufficient evidence is reached for anticipated effect or for futility when the boundary is crossed [27]. In this study, the assumptions included two-sided testing, type I error of 5% and power of 80%. Diversity-adjusted information size was calculated based on a control event rates (the absolute AKI incidence in the control arms of included trials) and relative risk reduction of 20%. The 20% relative risk reduction was chosen because it is an effect size that is both clinically meaningful and realistic in cardiovascular trials. Boundaries for concluding superiority or inferiority or futility were calculated with the O'Brien–Fleming α-spending function. This analysis was carried out using a specific software (User Manual for TSA; Copenhagen Trial Unit 2011, Denmark). Results Search results and study characteristics Fig. 1 is a flow diagram of the study and summarizes the process of RCT selection. Eight eligible RCTs were identified, which enrolled 3,204 patients randomly assigned to receive statin therapy or placebo/no intervention. The earliest study was published in 1999 and the 3 most recent studies were published in 2016. Study sample sizes varied from 40 to 1,922. Main characteristics of included trials on study design, surgery type, patient characteristic, statin therapy (varied widely in terms of drug, dose, timing of administration, and duration) and AKI definition is listed in Table 1.
Page 7
Assessment of risk of bias Risk of bias assessment found 4 studies with low risk of bias. One study was considered to have high risk of bias because the authors did not clearly address reporting of incomplete outcome data. Other 3 studies had unclear risk of bias because random sequence generation and/or allocation concealment were not clearly described (Fig. 2). Primary outcome: incidence of AKI Postoperative AKI was reported in all included studies, although different definitions were applied by the investigators of respective trials, including the AKIN criterion [22-24], RIFLE criterion [21], and others [17-20]. Postoperative AKI occurred in 341 of the 1,603 patients receiving statin therapy and in 301 of the 1,601 patients in the control arms. Analyzed with a random-effects model, the pooled estimate of the eight trials suggested that perioperative statin therapy was not associated with a significant reduction in the incidence of postoperative AKI (RR=1.02, 95% CI=0.82-1.28, P=0.85; heterogeneity I2=26%; Fig. 2). Visual inspection of the funnel plot showed some asymmetry (Supplementary Fig. 1), but we did not conduct statistical tests for the asymmetry due to the small number of included trials [25]. Various sensitivity analyses were performed for the primary outcome. Meta-analysis using a fixed-effects model did not show significantly different AKI incidence in statin therapy arms and in control arms. Besides, meta-analyses were repeated after excluding studies with unclear and high risk of bias, excluding studies that used statins other than atorvastatin, excluding studies that did not report AKI as a primary outcome, excluding studies that did not use RIFLE or AKIN criteria for AKI diagnosis, and excluding patients that were chronic statin users. None of these found a significant effect of statin therapy on
Page 8
AKI incidence (Supplementary Table 2). Furthermore, sensitivity analysis by omitting one trial at a time showed that none of the included trials markedly changed the overall result. We conducted trial sequential analysis of statin therapy versus control on postoperative AKI incidence (Fig. 3). Based on a control event rate of 18.8% (301 events in 1,601 patients) and an a priori anticipated intervention effect of a relative risk reduction of 20 %, trial sequential analysis estimated that the optimal sample size to draw a conclusion was 8,746. In the present meta-analysis, the number of patients enrolled in all trials was 3,204 and the overall estimate was not statistically significant. The cumulative Z curve breaks through the boundary for futility (nonsuperiority). Thus, an intervention effect of 20 % relative risk reduction or more on AKI incidence is unlikely. Secondary outcomes The results of analyses on secondary outcomes are summarized in Supplementary Table 3. These outcomes were only reported in some of the included studies. Compared with control, perioperative statin therapy did not significantly alter the clinical outcomes of the need for renal replacement therapy (RR=1.09, 95% CI=0.45-2.66; number of studies n=3), mechanical ventilation duration (MD=24.84 min, 95% CI=-55.53-105.20; n=2), ICU length of stay (MD=0.04 days, 95% CI=-3.13-3.20, n=7), hospital length of stay (MD=-0.08 days, 95% CI=-0.31-0.15; n=6), or in-hospital mortality (RR=3.76, 95% CI=0.93-15.14; n=7). Discussion In this meta-analysis consisting of eight RCTs that enrolled a total of 3,204 patients undergoing open-heart surgery, we found that perioperative statin therapy did not significantly affect the incidence of AKI after surgery. Trial sequential analysis suggested that sufficient evidence had accrued to reject the possibility that perioperative statin therapy could achieve a 20 % or more relative risk reduction in AKI incidence. Statin had also failed
Page 9
in improving other postoperative outcomes including the need for renal replacement therapy, mechanical ventilation duration, length of stays in ICU and hospital, and in-hospital mortality. Previously, a Cochrane review had summarized RCTs that assessed the impact of preoperative statin use for surgery requiring cardiac bypass on the incidence of postoperative AKI [28]. It showed that the risk of AKI was nominally lower in the statin group than the control group, but the benefit did not reach statistical significance (RR=0.76, 95% CI= 0.461.28). However, this study included only five small RCTs that enrolled 467 patients in total. Several other relevant review articles published in recent years included observational studies in their analyses [9-11, 14]. Many, but not all, of them showed that perioperative statin therapy was associated with a significant risk reduction in postoperative AKI. However, the results may have been biased because most patients included in these studies were from observational studies while data from RCTs only contributed a negligible part. Observational studies frequently reach distorted conclusions because they are influenced by confounding. Conclusive evidence on this clinical issue should be addressed by well-conducted large-scale randomized trials. Fortunately, with the recent publication of several RCTs, the information size regarding this issue has dramatically increased, making an updated meta-analysis warranted. The result of the present meta-analysis is consistent with that of the Cochrane review that perioperative statin therapy brings no benefit in reducing AKI following openheat surgery. Another meta-analysis of RCTs that included recently published trials and reported the outcome of AKI was noticed [29]. In that study, meta-analysis based primarily on three trials with low risk of bias reached a conclusion that statin therapy was associated with an increase in incidence of AKI after cardiac surgery. The result was driven by a highly influential trial, which got 70.2% of the weight [24]. However, this trial did not study postoperative AKI as a primary outcome, and the investigators did not control for multiple comparisons. Thus, the
Page 10
probability of incorrect rejections of the null hypotheses (type I errors) was high [30]. On the contrary, our meta-analysis including all available trials showed statin therapy did not change the risk of AKI. This is supported by additional analyses, including (1) the trial sequential analysis showed enough evidence had accumulated that statin is unlikely to influence the incidence of AKI, (2) various sensitivity analyses of trials/patients with different characteristics showed non-significant results, and (3) other outcomes related with AKI, such as need for renal replacement therapy, length of stay and mortality, were unchanged by the statin therapy. Statins are commonly used lipid-lowering drugs, but their non-lipid-lowering activities, such as anti-inflammatory, antioxidative, and endothelial protecting effects, made them a potential agent for protecting renal function [31, 32]. There is growing evidence supporting the effectiveness of statins in reducing contrast-induced AKI in patients undergoing diagnostic or interventional procedures with contrast media [33, 34]. In cardiac surgery, however, statins’ renal protecting effect has only been shown in some observational studies [35, 36]. None of the three RCTs that studied AKI as a primary outcome demonstrated a significantly reduced postoperative AKI incidence in the statin arms [22-24], and our present meta-analysis of eight RCTs has confirmed this conclusion. The reason why statins can prevent contrast-induced AKI but not cardiac surgery-associated AKI is unknown. Similar observations have also been made in trials of N-acetylcysteine and sodium bicarbonate. Although some uncertainty remains [37, 38], a number of randomized trials as well as metaanalyses support the effectiveness of N-acetylcysteine and sodium bicarbonate therapy for the prevention of contrast-induced AKI [39, 40, 41, 42]. However, both of them have failed to prevent AKI after open-heart surgery [43, 44]. It is possible that the discrepancy in prophylactic efficacies is related to the different degree of inflammatory reaction in the two clinical settings. These medications alone might not be sufficient to protect the kidney against
Page 11
marked systemic inflammation and oxidative stress induced in heart surgery. Another possible reason is that some type of statins, if not all, might have adverse effects on renal function. For example, rosuvastatin treatment has been associated with new proteinuria onset [45].These adverse effects might have offset the potential renoprotective effects of statins through their anti-inflammatory and antioxidant activities. In addition, some authors suggested that statin-naive patients, who tend to have less cardiovascular morbidity, might get limited benefit from statins [23]. Our results should be interpreted in light of some limitations. First of all, the definitions of AKI were different across included studies. Some used serum creatinine cut-offs to dichotomize patients into AKI or no AKI. However, when we excluded studies that did not use the currently recommended criteria (RIFLE or AKIN) in a sensitivity analysis, the results remained unchanged. Second, the amount of evidence concerning the secondary outcomes such as the need for renal replacement is small and firm conclusions could not be reached. Third, the statin therapy used in individual studies varied widely in terms of drug, dose, timing of administration, and duration. We could not perform subgroup analysis on these factors due to the limited number of studies. Furthermore, the safety and efficacy of prophylactic statins might be different in specific patient subgroups, such as long-term statin users, those with chronic kidney disease or diabetes, and patients receiving on-pump or offpump surgeries. In this study-level meta-analysis, we could not examine the effects of these variables. The renal effects of perioperative statin therapy should be further studied for specific statin regimens and patient groups. We searched the ClincalTrials.gov website and noticed some ongoing trials on statin treatment in cardiac surgery. For example, the StaRT-CABG trial (NCT01715714) plans to enroll 2,630 patients to study the effects of high doses atorvastatin before coronary artery bypass grafting in chronic statin users. The results of these
Page 12
trials will answer part of this question. Besides, the subgroup analyses of trials are usually underpowered to identify subgroup effects. A meta-analysis of individual patient data from existing clinical trials would provide greater statistical power [46]. Such analysis may shed more light on the true effects of statin treatment in specific patient and care-delivery subgroups. In conclusion, in this updated meta-analysis with trial sequential analysis of RCTs of patients undergoing cardiac surgery, perioperative statin treatment did not reduce the risk of AKI. The current evidence does not support the use of statin therapy to prevent AKI following cardiac surgery.
Page 13
References [1] Rosner MH, Okusa MD. Acute kidney injury associated with cardiac surgery. Clin J Am Soc Nephrol 2006;1:19-32. [2] Siddiqui NF, Coca SG, Devereaux PJ, Jain AK, Li L, Luo J, et al. Secular trends in acute dialysis after elective major surgery-1995 to 2009. CMAJ 2012;184:1237-45. [3] Birnie K, Verheyden V, Pagano D, Bhabra M, Tilling K, Sterne JA, et al. Predictive models for kidney disease: improving global outcomes (KDIGO) defined acute kidney injury in UK cardiac surgery. Crit Care 2014;18:606. [4] Thakar CV, Yared JP, Worley S, Cotman K, Paganini EP. Renal dysfunction and serious infections after open-heart surgery. Kidney Int 2003;64:239-46. [5] Hoste EA, Bagshaw SM, Bellomo R, Cely CM, Colman R, Cruz DN, et al. Epidemiology of acute kidney injury in critically ill patients: the multinational AKI-EPI study. Intensive Care Med 2015;41:1411-23. [6] Lafrance JP, Miller DR. Acute kidney injury associates with increased long-term mortality. J Am Soc Nephrol 2010;21:345-52. [7] Pickering JW, James MT, Palmer SC. Acute kidney injury and prognosis after cardiopulmonary bypass: a meta-analysis of cohort studies. Am J Kidney Dis 2015;65:283-93. [8] Liakopoulos OJ, Choi YH, Haldenwang PL, Strauch J, Wittwer T, Dorge H, et al. Impact of preoperative statin therapy on adverse postoperative outcomes in patients undergoing cardiac surgery: a meta-analysis of over 30,000 patients. Eur Heart J 2008;29:1548-59. [9] Wang J, Gu C, Gao M, Yu W, Yu Y. Preoperative Statin Therapy and Renal Outcomes After Cardiac Surgery: A Meta-analysis and Meta-regression of 59,771 Patients. Can J Cardiol 2015;31:1051-60.
Page 14
[10] Pan SY, Wu VC, Huang TM, Chou HC, Ko WJ, Wu KD, et al. Effect of preoperative statin therapy on postoperative acute kidney injury in patients undergoing major surgery: systemic review and meta-analysis. Nephrology (Carlton) 2014;19:750-63. [11] Kuhn EW, Liakopoulos OJ, Stange S, Deppe AC, Slottosch I, Choi YH, et al. Preoperative statin therapy in cardiac surgery: a meta-analysis of 90,000 patients. Eur J Cardiothorac Surg 2014;45:17-26; discussion [12] Singh I, Rajagopalan S, Srinivasan A, Achuthan S, Dhamija P, Hota D, et al. Preoperative statin therapy is associated with lower requirement of renal replacement therapy in patients undergoing cardiac surgery: a meta-analysis of observational studies. Interact Cardiovasc Thorac Surg 2013;17:345-52. [13] Kuhn EW, Slottosch I, Wahlers T, Liakopoulos OJ. Preoperative statin therapy for patients undergoing cardiac surgery. Cochrane Database Syst Rev 2015:CD008493. [14] Li M, Zou H, Xu G. The prevention of statins against AKI and mortality following cardiac surgery: A meta-analysis. Int J Cardiol 2016;222:260-6. [15] Hillis LD, Smith PK, Anderson JL, Bittl JA, Bridges CR, Byrne JG, et al. 2011 ACCF/AHA Guideline for Coronary Artery Bypass Graft Surgery. A report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. Developed in collaboration with the American Association for Thoracic Surgery, Society of Cardiovascular Anesthesiologists, and Society of Thoracic Surgeons. J Am Coll Cardiol 2011;58:e123-210. [16] Authors/Task Force members, Windecker S, Kolh P, Alfonso F, Collet JP, Cremer J, et al. 2014 ESC/EACTS Guidelines on myocardial revascularization: The Task Force on Myocardial Revascularization of the European Society of Cardiology (ESC) and the European Association for Cardio-Thoracic Surgery (EACTS)Developed with the special
Page 15
contribution of the European Association of Percutaneous Cardiovascular Interventions (EAPCI). Eur Heart J 2014;35:2541-619. [17] Christenson JT. Preoperative lipid-control with simvastatin reduces the risk of postoperative thrombocytosis and thrombotic complications following CABG. Eur J Cardiothorac Surg 1999;15:394-9; discussion 9-400. [18] Spadaccio C, Pollari F, Casacalenda A, Alfano G, Genovese J, Covino E, et al. Atorvastatin increases the number of endothelial progenitor cells after cardiac surgery: a randomized control study. J Cardiovasc Pharmacol 2010;55:30-8. [19] Chello M, Patti G, Candura D, Mastrobuoni S, Di Sciascio G, Agro F, et al. Effects of atorvastatin on systemic inflammatory response after coronary bypass surgery. Crit Care Med 2006;34:660-7. [20] Mannacio VA, Iorio D, De Amicis V, Di Lello F, Musumeci F. Effect of rosuvastatin pretreatment on myocardial damage after coronary surgery: a randomized trial. J Thorac Cardiovasc Surg 2008;136:1541-8. [21] Prowle JR, Calzavacca P, Licari E, Ligabo EV, Echeverri JE, Haase M, et al. Pilot double-blind, randomized controlled trial of short-term atorvastatin for prevention of acute kidney injury after cardiac surgery. Nephrology (Carlton) 2012;17:215-24. [22] Billings FTt, Hendricks PA, Schildcrout JS, Shi Y, Petracek MR, Byrne JG, et al. Highdose perioperative atorvastatin and acute kidney injury following cardiac surgery: A randomized clinical trial. JAMA 2016;315:877-88. [23] Park JH, Shim JK, Song JW, Soh S, Kwak YL. Effect of atorvastatin on the incidence of acute kidney injury following valvular heart surgery: a randomized, placebo-controlled trial. Intensive Care Med 2016;42:1398-407. [24] Zheng Z, Jayaram R, Jiang L, Emberson J, Zhao Y, Li Q, et al. Perioperative rosuvastatin in cardiac surgery. N Engl J Med 2016;374:1744-53.
Page 16
[25] Higgins J, Green S. Cochrane handbook for systematic reviews of interventions, Version 5.1.0 (Updated March 2011), http://www.cochrane-handbook.org (accessed September 12, 2016) [26] Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 2009;6:e1000097. . [27] Wetterslev J, Thorlund K, Brok J, Gluud C. Trial sequential analysis may establish when firm evidence is reached in cumulative meta-analysis. J Clin Epidemiol 2008;61:64-75. [28] Lewicki M, Ng I, Schneider AG. HMG CoA reductase inhibitors (statins) for preventing acute kidney injury after surgical procedures requiring cardiac bypass. Cochrane Database Syst Rev 2015:CD010480. [29] Putzu A, Capelli B, Belletti A, et al. Perioperative statin therapy in cardiac surgery: a meta-analysis of randomized controlled trials. Crit Care 2016; 20:395. [30] Aickin M, Gensler H. Adjusting for multiple testing when reporting research results: the Bonferroni vs Holm methods. Am J Public Health 1996;86:726-8. [31] Epstein M, Campese VM. Pleiotropic effects of 3-hydroxy-3-methylglutaryl coenzyme a reductase inhibitors on renal function. Am J Kidney Dis 2005;45:2-14. [32] Campese VM, Nadim MK, Epstein M. Are 3-hydroxy-3-methylglutaryl-CoA reductase inhibitors renoprotective? J Am Soc Nephrol 2005;16 Suppl 1:S11-7. [33] Jo SH, Koo BK, Park JS, Kang HJ, Cho YS, Kim YJ, et al. Prevention of radiocontrast medium-induced nephropathy using short-term high-dose simvastatin in patients with renal insufficiency undergoing coronary angiography (PROMISS) trial - a randomized controlled study. Am Heart J 2008;155. [34] Su X, Xie X, Liu L, Lv J, Song F, Perkovic V, et al. Comparative effectiveness of 12 treatment strategies for preventing contrast-induced acute kidney injury: A systematic review and Bayesian network meta-analysis. Am J Kidney Dis 2017;69:69-77.
Page 17
[35] Layton JB, Kshirsagar AV, Simpson RJ, Pate V, Funk MJ, Sturmer T, et al. Effect of statin use on acute kidney injury risk following coronary artery bypass grafting. Am J Cardiol 2013;111:823-8. [36] Layton JB, Hansen MK, Jakobsen CJ, Kshirsagar AV, Andreasen JJ, Hjortdal VE, et al. Statin initiation and acute kidney injury following elective cardiovascular surgery: a population cohort study in Denmark. Eur J Cardio-Thoracic Surg 2016;49:995-1000. [37] ACT Investigators. Acetylcysteine for prevention of renal outcomes in patients undergoing coronary and peripheral vascular angiography: main results from the randomized Acetylcysteine for Contrast-induced nephropathy Trial (ACT). Circulation. 2011;124:1250-9. [38] Lee SW, Kim WJ, Kim YH, Park SW, Park DW, Yun SC, et al. Preventive strategies of renal insufficiency in patients with diabetes undergoing intervention or arteriography (the PREVENT trial). Am J Cardiol 2011;107:1447-52. [39] Marenzi G, Assanelli E, Marana I, Lauri G, Campodonico J, Grazi M, et al. Nacetylcysteine and contrast-induced nephropathy in primary angioplasty. N Engl J Med 2006;354:2773-82. [40] Merten GJ, Burgess WP, Gray LV, Holleman JH, Roush TS, Kowalchuk GJ, et al. Prevention of contrast-induced nephropathy with sodium bicarbonate: a randomized controlled trial. JAMA 2004;291:2328-34. [41] Xu R, Tao A, Bai Y, Deng Y, Chen G. Effectiveness of N-acetylcysteine for the prevention of contrast-induced nephropathy: A systematic review and meta-analysis of randomized controlled trials. J Am Heart Assoc 2016;5: e003968. [42] Jang JS, Jin HY, Seo JS, Yang TH, Kim DK, Kim TH, et al. Sodium bicarbonate therapy for the prevention of contrast-induced acute kidney injury – a systematic review and metaanalysis –. Circ J 2012;76:2255-65.
Page 18
[43] Adabag AS, Ishani A, Koneswaran S, Johnson DJ, Kelly RF, Ward HB, et al. Utility of N-acetylcysteine to prevent acute kidney injury after cardiac surgery: a randomized controlled trial. Am Heart J 2008;155:1143-9. [44] Haase M, Haase-Fielitz A, Plass M, Kuppe H, Hetzer R, Hannon C, et al. Prophylactic perioperative sodium bicarbonate to prevent acute kidney injury following open heart surgery: a multicenter double-blinded randomized controlled trial. PLoS Med 2013;10:e1001426. [45] Savarese G, Musella F, Volpe M, Paneni F, Perrone-Filardi P. Effects of atorvastatin and rosuvastatin on renal function: a meta-analysis. Int J Cardiol 2013;167:2482-9. [46] Reade MC, Delaney A, Bailey MJ, Harrison DA, Yealy DM, Jones PG, et al. Prospective meta-analysis using individual patient data in intensive care medicine. Intensive Care Med 2010;36:11-21.
Fig. 1. PRISMA flow diagram of the study. Fig. 2. Meta-analysis of perioperative statins versus control for acute kidney injury prevention: forest plot of the risk ratio of acute kidney injury in patients treated with statins compared with patients treated with control; risk of bias assessment of included studies. Fig. 3. Trial sequential analysis of included studies with a control event rate of 18.8%, type I error of 5%, power of 80%, and anticipated relative risk reduction of 20%.
Page 19
Table 1. Basic characteristics of randomized trials included in meta-analysis Author, Study Surgery CPB Age Sex Chronic Statins type, dosage, year, use (yrs) (M, %) statins design type timing reference (%) use (%) Christenson OL, R CABG 1999 [17]
100 63
80.5
NR
simvastatin 20 mg/d for 4w before surgery
Chello 2006 [19] Mannacio 2008 [20]
DB, CABG PC, R DB, CABG PC, R
100 65
77.5
0
100 60
72.5
NR
atorvastatin 20 mg/d for 3w before surgery rosuvastatin 20 mg/d for 7d before surgery
Spadaccio 2010 [18] Prowle 2012 [21]
DB, CABG 100 65 PC, R DB, CABG, 100 68 PC, R valvular
54.0
0
70.0
70
Billings DB, CABG, 70.8 67 2016 [22] PC, R valvular
69.4
67.6
Park 2016 DB, valvular 100 58 [23] PC, R
49.5
0
Zheng DB, CABG, 53.2 59 2016 [24] PC, R valvular
79.2
34.0
atorvastatin 20 mg/d for 3w before surgery atorvastatin 40 mg the morning of surgery and on postoperative days 1-3 atorvastatin 80 mg the day before surgery, 40 mg the morning of surgery, and 40 mg/d after surgery until discharge atorvastatin 80 mg the evening before surgery, 40 mg the morning of surgery, and 40 mg/d on postoperative days 0-2 rosuvastatin 20 mg/d for up to 8 days before surgery and for 5 days thereafter
AKI definition serum urea ≥ 9 mmol/L and SCr ≥ 125 µmol/L NR increase in SCr > 221 µmol/L NR RIFLE category R or greater AKIN criteria stage 1 or greater
AKIN criteria stage 1 or greater
AKIN criteria stage 1 or greater
AKI = acute kidney injury, AKIN = Acute Kidney Injury Network, CABG = coronary artery bypass grafting, CPB = cardiopulmonary bypass, DB = double blind, M = male, NR = not reported, OL = open label, PC = placebo controlled, R = randomized, RIFLE = Risk of renal injury/Injury to the kidney/Failure of kidney function/Loss of kidney function/End stage disease, SCr = serum creatinine.
Page 20
Page 21