SYSTEMATIC REVIEW published: 08 June 2021 doi: 10.3389/fcvm.2021.676771 Efficacy and Safety of Colchicine in Post–acute Myocardial Infarction Patients: A Systematic Review and Meta-Analysis of Randomized Controlled Trials Carlos Diaz-Arocutipa 1,2*, Jerry K. Benites-Meza 3,4 , Diego Chambergo-Michilot 3,5 , Joshuan J. Barboza 1,3 , Vinay Pasupuleti 6 , Héctor Bueno 7,8,9,10 , Antonia Sambola 10,11 and Adrian V. Hernandez 1,12 1 Vicerrectorado de Investigación, Universidad San Ignacio de Loyola, Lima, Peru, 2 Programa de Atención Domiciliaria – EsSalud, Lima, Peru, 3 Tau Relaped Group, Trujillo, Peru, 4 Facultad de Medicina, Universidad Nacional de Trujillo, Trujillo, Peru, 5 Facultad de Ciencias de la Salud, Universidad Científica del Sur, Lima, Peru, 6 MedErgy HealthGroup, Inc., Yardley, PA, United States, 7 Centro Nacional de Investigaciones Cardiovasculares, Madrid, Spain, 8 Cardiology Department, Hospital Universitario 12 de Octubre, Instituto de Investigación Sanitaria Hospital 12 de Octubre (imas12), Madrid, Spain, 9 Facultad de Medicina, Universidad Complutense de Madrid, Madrid, Spain, 10 Centro de Investigación Biomédica en Red Enfermedades Cardiovasculares, Madrid, Spain, 11 Department of Cardiology, University Hospital Vall d’hebron, Universitat Edited by: Autònoma, Barcelona, Spain, 12 Health Outcomes, Policy, and Evidence Synthesis Group, University of Connecticut School Turgay Celik, of Pharmacy, Storrs, CT, United States VM Medical Park Ankara (Kecioren), Turkey Reviewed by: Background: Inflammation plays a key role in atherosclerotic plaque destabilization and Gianluca Campo, adverse cardiac remodeling. Recent evidence has shown a promising role of colchicine in University Hospital of Ferrara, Italy Matthias Bossard, patients with coronary artery disease. We evaluated the efficacy and safety of colchicine Luzerner Kantonsspital, Switzerland in post–acute myocardial infarction (MI) patients. *Correspondence: Methods: We searched five electronic databases from inception to January 18, Carlos Diaz-Arocutipa carlosdiaz013@gmail.com 2021, for randomized controlled trials (RCTs) evaluating colchicine in post–acute MI patients. Primary outcomes were cardiovascular mortality and recurrent MI. Secondary Specialty section: outcomes were all-cause mortality, stroke, urgent coronary revascularization, levels of This article was submitted to Coronary Artery Disease, follow-up high-sensitivity C-reactive protein (hs-CRP), and drug-related adverse events. a section of the journal All meta-analyses used inverse-variance random-effects models. Frontiers in Cardiovascular Medicine Results: Six RCTs involving 6,005 patients were included. Colchicine did not significantly Received: 06 March 2021 Accepted: 22 April 2021 reduce cardiovascular mortality [risk ratio (RR), 0.91; 95% confidence interval (95% Published: 08 June 2021 CI), 0.52–1.61; p = 0.64], recurrent MI (RR, 0.87; 95% CI, 0.62–1.22; p = 0.28), Citation: all-cause mortality (RR, 1.06; 95% CI, 0.61–1.85; p = 0.78), stroke (RR, 0.28; 95% CI, Diaz-Arocutipa C, Benites-Meza JK, Chambergo-Michilot D, Barboza JJ, 0.07–1.09; p = 0.05), urgent coronary revascularization (RR, 0.46; 95% CI, 0.02–8.89; Pasupuleti V, Bueno H, Sambola A p = 0.19), or decreased levels of follow-up hs-CRP (mean difference, −1.95 mg/L; and Hernandez AV (2021) Efficacy and 95% CI, −12.88 to 8.98; p = 0.61) compared to the control group. There was Safety of Colchicine in Post–acute Myocardial Infarction Patients: A no increase in any adverse events (RR, 0.97; 95% CI, 0.89–1.07; p = 0.34) or Systematic Review and Meta-Analysis gastrointestinal adverse events (RR, 2.49; 95% CI, 0.48–12.99; p = 0.20). Subgroup of Randomized Controlled Trials. Front. Cardiovasc. Med. 8:676771. analyses by colchicine dose (0.5 vs. 1 mg/day), time of follow-up (<1 vs. ≥1 year), doi: 10.3389/fcvm.2021.676771 and treatment duration (≤30 vs. >30 days) showed no changes in the overall findings. Frontiers in Cardiovascular Medicine | www.frontiersin.org 1 June 2021 | Volume 8 | Article 676771 Diaz-Arocutipa et al. Colchicine in Acute Myocardial Infarction Conclusion: In post–acute MI patients, colchicine does not reduce cardiovascular or all-cause mortality, recurrent MI, or other cardiovascular outcomes. Also, colchicine did not increase drug-related adverse events. Keywords: colchicine, myocardial infarction, atherosclerosis, inflammation, meta-analysis INTRODUCTION applied the Cochrane highly sensitive search strategy for identifying randomized controlled trials (RCTs) in PubMed Atherosclerotic coronary artery disease (CAD) remains a leading (13). There were no restrictions on language or publication cause of global morbidity and mortality (1), despite progress date. Also, we conducted a hand search of reference lists in medical and invasive treatment. There is a current interest of included studies and relevant review articles to further in inflammation as a therapeutic target in patients with identify eligible trials. Additionally, we searched trial registries acute coronary syndrome (ACS) as it plays a central role in www.clinicaltrials.gov/ and www.who.int/trialsearch/ for the destabilization of atherosclerotic plaques and ventricular completed as well as ongoing RCTs. remodeling after ACS (2). In recent years, some drugs with anti- inflammatory effects have been assessed in patients with CAD. Eligibility Criteria Canakinumab, a monoclonal antibody that inhibits interleukin Study inclusion criteria were as follows: (i) RCTs, (ii) adult 1β (IL-1β), significantly reduced major cardiovascular events and patients (≥18 years of age) post–acute MI defined according high-sensitivity C-reactive protein (hs-CRP) levels in patients to study authors, (iii) any dose and duration of colchicine as with a history of myocardial infarction (MI) and hs-CRP > experimental group, and (iv) placebo or standard treatment as 2 mg/L (3). However, canakinumab was associated with a a control group. We excluded observational studies, systematic higher incidence of fatal infections compared to placebo. In reviews, narrative reviews, editorials, letters to the editor, contrast, methotrexate did not show a significant reduction of and abstracts. cardiovascular events or hs-CRP levels in patients with a previous CAD (4). Given these discordant results, it was necessary Study Selection to search for new anti-inflammatory drugs with an adequate Two review authors (CDA and JJB) downloaded all titles and safety profile. abstracts from electronic search to EndNote X8 software and Colchicine is a low-cost drug with wider anti-inflammatory duplicate records were removed. Titles and abstracts of the properties that has been used for more than 200 years for the retrieved articles were independently screened by two authors treatment of gout attacks and currently in other inflammatory (CDA and JBM) to identify potentially relevant studies. Two diseases such as acute pericarditis and familial Mediterranean authors (CDA and JBM) independently screened the full text and fever. Few systematic reviews have examined the effect of registered reasons for the exclusion of studies. Any disagreement colchicine in patients with CAD (5–11), but none focused on on title/abstract and full-text selection was resolved by consensus. post–acute MI patients. The analysis of its effect specifically on MI patients is important because of the intense inflammatory Outcomes response associated with acute MI and the high risk for poor Primary outcomes were cardiovascular mortality and recurrent outcomes in these patients in whom the anti-inflammatory effect MI. Secondary outcomes were all-cause mortality, stroke, urgent of colchicine may be more relevant. We conducted a systematic coronary revascularization, follow-up hs-CRP level, any adverse review and meta-analysis to assess the efficacy and safety of events, and gastrointestinal adverse events. We used author- colchicine in post–acute MI patients. reported definitions for all outcomes. Data Extraction METHODS Information from each study was independently extracted by two authors (CDA and DCM) using a standardized data This review was reported according to the 2009 PRISMA extraction form that was previously piloted. Any disagreement (Preferred Reporting Items for Systematic Reviews and Meta- was resolved by consensus. If additional data were needed, we Analyses) statement (12) and was registered in the PROSPERO contacted the corresponding author through email to request database (CRD42020177536). further information. We extracted the following data: first author name, year of publication, country, type of RCT, sample size, Search Strategy population, age, sex, duration and dosage of colchicine, and We searched the following bibliographic databases: PubMed, primary and secondary outcomes. EMBASE, Scopus, Web of Science, and Cochrane Central Register of Controlled Trials. Searches were conducted from Risk-of-Bias Assessment inception to April 4, 2020, with an update on January 18, Two review authors (CDA and VP) independently assessed the 2021. The search strategy for PubMed was adapted for risk of bias of each study using the Cochrane Risk of Bias 2.0 tool use in the other databases (Supplementary Table 1). We (13). This tool evaluates five domains: randomization process, Frontiers in Cardiovascular Medicine | www.frontiersin.org 2 June 2021 | Volume 8 | Article 676771 Diaz-Arocutipa et al. Colchicine in Acute Myocardial Infarction deviations from intended interventions, missing outcome data, similar in three trials (20, 21, 23). The sample size (32–4,745 measurement of the outcome, and selection of the reported result. patients) and the length of follow-up (5 days−22.6 months) Overall, each RCT was judged as having a low, some concerns, were heterogeneous across studies. Four trials (20–23) included or a high risk of bias. Any disagreement was resolved by a third patients with non–ST-segment elevation MI (NSTEMI) and ST- author (AVH). segment elevation MI (STEMI), whereas in the other two trials (18, 19), only STEMI patients were enrolled. Four RCTs were GRADE Quality of Evidence double-blinded (19–21, 23), one RCT was open-label (18), and The Grading of Recommendations Assessment, Development, in one, RCT blinding was not specified (22). and Evaluation (GRADE) approach (14) was used to appraise Control groups were placebo or standard treatment across the quality of evidence for all outcomes. In using GRADE trials. Doses and treatment duration of colchicine varied among methodology, we considered the following criteria to assess the studies. The doses used were 0.5 (21, 22) and 1 (18–20) mg/day quality of evidence: risk of bias, inconsistency, indirectness, across five trials with a duration that ranged from 5 days to 24 imprecision, and publication bias. We generate the Summary of months. One trial combined both doses (Table 1) (23). Overall, Findings table using the GRADEpro software. discontinuation rate of colchicine across trials ranged from 2 to 26% (Table 1). Our search for ongoing trials identified three Statistical Analysis registered RCTs that assess the effect of colchicine in post–acute All meta-analyses were conducted using the inverse-variance MI patients (Supplementary Table 2). method and random-effects models with Hartung–Knapp adjustment given the small number of included trials (15). The between-study variance (τ ²) was estimated using the Paule– Risk-of-Bias Assessment Mandel estimator (16). We pooled risk ratios (RRs) and mean Three trials demonstrated an overall low risk of bias. The differences (MDs) with their 95% confidence intervals (95% CIs) other three trials showed some concerns because they reported for binary and continuous outcomes, respectively. Heterogeneity insufficient information regarding the randomization process among studies was evaluated using the χ 2 test (p < 0.10 was (Figure 2). considered as the presence of heterogeneity) and the I 2 statistic. Heterogeneity was defined as low if I 2 < 30%, moderate if I 2 = 30–60%, and high if I 2 > 60%. As the number of trials was <10, Primary Outcomes it was not possible to evaluate small-study effects (13). In the case In four RCTs including 5,821 patients, colchicine did not studies that reported only median and interquartile range, the significantly reduce the risk of cardiovascular mortality (0.79 mean and standard deviation were estimated using the method vs. 0.86%; RR, 0.91; 95% CI, 0.52–1.61; p = 0.64; I 2 = 0%) or by Wan et al. (17). A pre-specified subgroup analyses were recurrent MI (3.31 vs. 3.84%; RR, 0.87; 95% CI, 0.62–1.22; p = conducted according to (i) colchicine dose (0.5 vs. 1 mg/day), 0.28; I 2 = 0%) in comparison to control group (Figure 3). (ii) time of follow-up (<1 vs. ≥1 year), and (iii) treatment duration (≤30 vs. >30 days). In sensitivity analyses, we assessed (i) all meta-analyses performed without the Hartung–Knapp Secondary Outcomes adjustment and (ii) only studies with a low risk of bias. We used Colchicine did not significantly reduce the risk of all-cause the metabin and metacont functions from the meta package in R mortality (five RCTs; n = 5,972; 1.74 vs. 1.54%; RR, 1.06; 95% 3.6.3 (www.r-project.org) for all meta-analyses. A two-tailed p < CI, 0.61–1.85; p = 0.78; I 2 = 0%), stroke (two RCTs; n = 5,540; 0.05 was considered statistically significant. 0.25 vs. 0.90%; RR, 0.28; 95% CI, 0.07–1.09; p = 0.05; I 2 = 0%), urgent coronary revascularization (two RCTs; n = 5,540; 1.01 RESULTS vs. 2.23%; RR, 0.46; 95% CI, 0.02–8.89; p = 0.19; I 2 = 1%), or decreased levels of follow-up hs-CRP (four RCTs; n = 449; Study Selection MD, −1.95 mg/L; 95% CI, −12.88 to 8.98; p = 0.61; I 2 = 73%) Our search strategy identified initially 2,011 articles. After the compared to control group (Figure 4). When assessing sources removal of duplicates, 1,151 unique articles remained. After of high heterogeneity of follow-up hs-CRP levels, the exclusion of screening of studies by title and abstract, 1,081 articles were the trial by Deftereos et al. (19) found that the effect of colchicine excluded. After a full-text assessment of 70 articles, 64 articles on follow-up hs-CRP levels was similar to the overall analysis and were excluded: 32 trial registries, 17 commentaries, 8 other the heterogeneity was reduced (Supplementary Figure 1). populations, and 7 abstracts. Six RCTs were selected for analyses Colchicine did not increase the risk of any adverse events (Figure 1). (three RCTs; n = 5,777; 18.60 vs. 19.09%; RR, 0.97; 95% CI, 0.89–1.07; p = 0.34; I 2 = 0%) or gastrointestinal adverse events Study Characteristics (five RCTs; n = 5,972; 18.08 vs. 16.85%; RR, 2.49; 95% CI, 0.48– Main characteristics of the six included RCTs (n = 6,005) are 12.99; p = 0.20; I 2 = 72%) in comparison to the control group summarized in Table 1. Seventy-nine percent of patients were (Figure 4). The most frequently reported gastrointestinal adverse men, and the mean age ranged between 55.4 and 61 years. The events were nausea, vomiting, and diarrhea. The three trials proportion of patients with hypertension (46%) and diabetes (18, 19, 23) that provided information on myelotoxicity found (20%) was similar across all trials. Previous MI (15%) was also no cases in both groups. Frontiers in Cardiovascular Medicine | www.frontiersin.org 3 June 2021 | Volume 8 | Article 676771 Diaz-Arocutipa et al. Colchicine in Acute Myocardial Infarction FIGURE 1 | Flow diagram of study selection. Subgroup Analyses adjustment, colchicine reduced only the risk of stroke or urgent Subgroup analyses by colchicine dose (0.5 vs. 1 mg/day), coronary revascularization in comparison to the control group time of follow-up (<1 vs. ≥1 year), and treatment duration (Supplementary Table 6). In addition, when only trials with a (≤30 vs. >30 days) did not show differences with the overall low risk of bias were pooled, the results were similar to the overall analyses for cardiovascular mortality, recurrent MI, all-cause analysis (Supplementary Table 7) except for Deftereos et al. (19) mortality, follow-up levels of hs-CRP, any adverse events, and that reported significantly lower levels of hs-CRP in patients gastrointestinal adverse events (Supplementary Tables 3–5). The treated with colchicine compared to the control group. only exceptions were significantly lower risk of urgent coronary revascularization in the trials of Tardif et al. (21) (0.5 mg/day) GRADE Summary of Findings and Tong et al. (23) (1 mg/day) and a significant risk reduction Among the primary outcomes, cardiovascular mortality and of stroke with a dose of 0.5 mg/day reported in the trial by Tardif recurrent MI had low quality of evidence. There was also a low et al. (21) (Supplementary Table 3). quality of evidence for all-cause mortality and urgent coronary revascularization. Stroke and any adverse events had a moderate Sensitivity Analyses quality of evidence. Finally, gastrointestinal adverse events and The results of the sensitivity analyses showed that when the follow-up hs-CRP levels were scored as very low quality of meta-analyses were performed without the Hartung–Knapp evidence (Table 2). Frontiers in Cardiovascular Medicine | www.frontiersin.org 4 June 2021 | Volume 8 | Article 676771 Frontiers in Cardiovascular Medicine | www.frontiersin.org Diaz-Arocutipa et al. TABLE 1 | Main characteristics of the included studies. References Country Type of Population Colchicine Initiation of Control Time of Arm Sample Age (years) Male (%) Discon CV mortality Recurrent Follow-up RCT colchicine follow-up size tinuation MI hs-CRP (%) (mg/L) Akodad France Open-label STEMI patients 1 mg QD for Within first day Standard 30 days Colchicine 23 60.1 ± 13.1* 82% 13% 0/23 (0%) 0/23 (0%) 29 ± 25.6* et al. (18) with ≤12 h 30 days of acute MI treatment successfully treated with PCI Control 22 59.7 ± 11.4* 76% NR 0/21 (0%) 1/21 21.9 ± 25.4* (4.76%) Deftereos Greece Double- STEMI patients 2 mg loading After coronary Placebo 5 days Colchicine 77 58 (52–54)† 70% 26% NR NR 42.9 et al. (19) blinded with ≤12 h from dose, then angiography (16.3–71.4)† the onset of 0.5 mg BID chest pain for 5 days Control 74 58 (51–68)† 68% 4% NR NR 63.8 (34.7–103.4)† Hennessy Australia Double- Patients with 0.5 mg BID Within 7 Placebo 30 days Colchicine 119 61 ± 13.6* 75% 2% 0/119 (0%) 0/119 (0%) 1.6 (0.7–3.5)† et al. (20) blinded type 1 MI within for 30 days days of the prior 7 days acute MI Control 118 61 ± 12.5* 79% 4% 0/118 (0%) 2/118 2 (0.9–4)† (1.69%) 5 Tardif et al. 12 Double- Patients with MI 0.5 mg Within 30 days Placebo 22.6 months Colchicine 2,366 60.6 ± 10.7* 80% 18.4% 20/2,366 89/2,366 1.37 (21) Countries blinded < 30 days who QD for of MI (13.4 ± (median) (0.84%) (3.76%) (0.75–2.13)‡ had completed 24 months 10.2 days)* any planned PCI Control 2,379 60.5 ± 10.6* 82% 18.7% 24/2,379 98/2,379 1.6 (1%) (4.11%) (0.9–2.65)‡ Wasyanto Indonesia Blinding Patients with 0.5 mg Not reported Placebo 5 days Colchicine 16 57.9* 87% NR NR NR NR et al. (22) unspecified acute MI QD for 5 days Control 16 52.9* 87% NR NR NR NR Tong et al. Australia Double- Patients 0.5 mg BID After coronary Placebo 400 days Colchicine 396 59.7 ± 10.2* 81% 15% 3/396 7/396 NR (23) blinded presenting with for the first angiography (0.75%) (1.76%) ACS (96.7% month, Colchicine in Acute Myocardial Infarction STEMI or followed by June 2021 | Volume 8 | Article 676771 NSTEMI) with 0.5 mg BID evidence for 11 of CAD months Control 399 60 ± 10.4* 78% 8% 1/399 11/399 NR (0.25%) (2.75%) RCT, randomized controlled trial; STEMI, ST-segment elevation myocardial infarction; NSTEMI, non–ST-segment elevation myocardial infarction; MI, myocardial infarction; PCI, percutaneous coronary intervention; QD, once a day; BID, twice a day; ACS, acute coronary syndrome; CAD, coronary artery disease; CV, cardiovascular; GI, gastrointestinal; hs-CRP, high-sensitivity C-reactive protein; GI, gastrointestinal; NR, not reported. *Mean ± standard deviation. † Median (interquartile range). ‡ Geometric mean (interquartile range). Diaz-Arocutipa et al. Colchicine in Acute Myocardial Infarction FIGURE 2 | Risk of Bias 2.0 tool for risk-of-bias assessment of included randomized controlled trials. FIGURE 3 | Effect of colchicine vs. control on (A) cardiovascular mortality and (B) recurrent myocardial infarction. RR, risk ratio; CI, confidence interval. DISCUSSION recurrent MI, all-cause mortality, stroke, urgent coronary revascularization, or follow-up levels of hs-CRP compared to In our meta-analysis of six RCTs, we found that colchicine did placebo or standard treatment. Also, there was no increase in not significantly reduce the risk of cardiovascular mortality, any adverse events or gastrointestinal adverse events associated Frontiers in Cardiovascular Medicine | www.frontiersin.org 6 June 2021 | Volume 8 | Article 676771 Diaz-Arocutipa et al. Colchicine in Acute Myocardial Infarction FIGURE 4 | Effects of colchicine on (A) all-cause mortality, (B) stroke, (C) urgent coronary revascularization, (D) any adverse events, (E) gastrointestinal adverse events, and (F) follow-up levels of hs-CRP (mg/L). hs-CRP, high-sensitivity C-reactive protein; RR, risk ratio; CI, confidence interval; SD, standard deviation; MD, mean difference. with the use of colchicine. Overall, both the subgroup also decreased monocyte– and neutrophil–platelet aggregation analyses by colchicine dose, time of follow-up, or treatment and expression of surface markers of platelet activity, thus duration and the sensitivity analyses were consistent with the targeting the platelet-inflammatory axis (32). On the other main results. hand, preclinical studies based on animal models of acute Colchicine has gained recent attention as a promising MI showed that the administration of colchicine produced a anti-inflammatory therapy for patients with CAD. Over the reduction of infarct size and cardiac remodeling in the acute and last years, several trials evaluating the role of colchicine chronic phase post–myocardial injury, probably as a result of the as an additional treatment to standard medical therapy for inhibition of the inflammatory response (33–35). These findings secondary prevention in patients with chronic CAD or after were corroborated in a clinical trial of STEMI patients treated ACS have been published. Colchicine acts by blocking the with a percutaneous coronary intervention (PCI) that showed polymerization of microtubules affecting several functions a lower infarct size defined by cardiac magnetic resonance in of inflammatory cells (chemotaxis, adhesion, phagocytosis, the group allocated to receive high-dose colchicine compared and protein excretion), interferes the neutrophil-endothelial with placebo (19). Overall, these data suggested that colchicine interaction through inhibition of the E-selectin expression on may have a beneficial effect in the setting of remodeling after an endothelial cells, and blocks the activity of the Nod-like protein acute MI. receptor 3 inflammasome within monocytes and neutrophils Clinical evidence of the impact of colchicine on major adverse at different levels (24–26). In patients with ACS, colchicine cardiovascular outcomes has produced divergent results. The acutely suppressed the caspase-1 activity and the local production COLCOT trial published by Tardif et al. (21) enrolled the largest of proinflammatory interleukins (IL-6, IL-1β, and IL-18) and population and showed a significant reduction of the primary chemokines, which are key mediators in the inflammatory composite endpoint [hazard ratio (HR), 0.77; 95% CI, 0.61–0.96; pathway of the coronary atherothrombosis (27–29). Given the p = 0.02] in patients with recent MI treated with colchicine diffuse coronary vascular inflammation observed in patients compared to placebo. However, this finding was mainly driven with acute MI (30), the reduction of the proinflammatory by a decrease in stroke and urgent angina revascularization with stimulus could lead to the stabilization of both culprit and no effect on cardiovascular mortality (HR, 0.84; 95% CI, 0.46– non-culprit atherosclerotic plaques, decreasing the recurrence 1.52) or recurrent MI (HR, 0.91; 95% CI, 0.68–1.21). Given of future coronary events. Interestingly, a study in 80 patients that hs-CRP values were measured in only 4.4% of included with recent ACS found that 1-year treatment with low-dose patients, we decided not to include the hs-CRP values from this colchicine favorably modified coronary plaques assessed by RCT (21) in our meta-analysis. Recently, a subanalysis of the coronary computed tomography angiography in comparison COLCOT trial showed that the beneficial effect of colchicine with standard medical therapy, irrespective of the reduction was restricted to patients who started this drug within 3 days of low-density lipoprotein levels (31). In addition, colchicine post-MI (HR, 0.52; 95% CI, 0.32–0.84), suggesting a possible Frontiers in Cardiovascular Medicine | www.frontiersin.org 7 June 2021 | Volume 8 | Article 676771 Diaz-Arocutipa et al. Colchicine in Acute Myocardial Infarction TABLE 2 | GRADE summary of findings. Outcomes Anticipated absolute effects* (95% CI) Relative effect No. of participants Certainty of the (95% CI) (studies) evidence (GRADE) Risk with Risk with control colchicine Cardiovascular mortality follow up: 9 per 1,000 8 per 1,000 RR, 0.91 5,821 ⊕⊕ range, 30 days−22.6 months (4–14) (0.52–1.61) (4 RCTs) LOWa,b Recurrent myocardial infarction 38 per 1,000 33 per 1,000 RR, 0.87 5,821 ⊕⊕ follow up: range, 30 days−22.6 (24–47) (0.62–1.22) (4 RCTs) LOWa,b months All-cause mortality follow up: range, 15 per 1,000 16 per 1,000 RR, 1.06 5,972 ⊕⊕ 5 days−22.6 months (9–28) (0.61–1.85) (5 RCTs) LOWb,c Stroke follow up: range, 400 9 per 1,000 3 per 1,000 RR, 0.28 5,540 ⊕⊕⊕ days−22.6 months (1–10) (0.07–1.09) (2 RCTs) MODERATEb Urgent coronary revascularization 22 per 1,000 10 per 1,000 RR, 0.46 5,540 ⊕⊕ follow up: range, 400 days−22.6 (0–198) (0.02–8.89) (2 RCTs) LOWb months High-sensitivity C-reactive protein — MD 1.95 mg/L — 449 ⊕ follow up: range, 5–30 days lower (4 RCTs) VERY LOWb,d,e (12.88 lower to 8.98 higher) Any adverse events follow up: range, 191 per 1,000 185 per 1,000 RR, 0.97 5,777 ⊕⊕⊕ 30 days−22.6 months (170–204) (0.89–1.07) (3 RCTs) MODERATEf Gastrointestinal adverse events 169 per 1,000 420 per 1,000 RR, 2.49 5,972 ⊕ follow up: range, 5 days−22.6 (81–1,000) (0.48–12.99) (5 RCTs) VERY LOWb,c,g months *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI, confidence interval; RR, risk ratio; MD, mean difference. The combination of symbols ⊕ and means the degree of certainty of the evidence as follows: very low (⊕ ), low (⊕ ⊕ ), moderate (⊕ ⊕ ⊕ ), and high (⊕ ⊕ ⊕⊕). GRADE Working Group grades of evidence: High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect. a Two of four studies had some concerns as risk of bias. b Wide confidence intervals. c Two of five studies had some concerns as risk of bias. d Three of four studies had some concerns as risk of bias. e I2 = 73%. f One of three studies had some concerns as risk of bias. g Outcome definition was heterogeneous across studies. benefit from early initiation of colchicine when inflammation caution with the interpretation of CIs is required when there are is more intense (36). In our review, the timing of colchicine RCTs of very unequal sizes as in our review (15). Shah et al. (37) initiation in three RCTs (18, 19, 23) was in the acute phase of reported that a single dose of oral pre-procedural colchicine did MI. In contrast, in the LoDoCo-MI (20) and COLCOT trials, not reduce target vessel revascularization myocardial injury or the initiation was within 7 and 30 days (mean of 13.4 ± 10.2 30-day major adverse cardiovascular events, but it did reduce days) after the index event, respectively. Because of the lack of post-procedure hs-CRP levels. This study was excluded from accurate reporting of treatment onset, a subgroup analysis was our review because the population was composed of patients not possible. The COPS trial published by Tong et al. (23) showed with CAD without adequate stratified randomization for the that colchicine did not significantly decrease cardiovascular subgroup with acute MI. Importantly, the treatment duration of outcomes at 400 days in patients with ACS. In contrast, an colchicine was heterogeneous across trials. In four of six RCTs increased risk of all-cause mortality (HR, 8.20; 95% CI, 1.03– (18, 20, 21, 23), the duration was 30 days or shorter, whereas in 65.61) was reported in the colchicine group compared to the the remaining two RCTs (19, 22), it was 12 months or longer. The control group. The COLCOT and COPS trials were the only ones inhomogeneity in the initiation and duration of colchicine in our to report the risk of stroke and urgent coronary revascularization. review raises the need that adequately powered RCTs with earlier Although both RCTs reported individual favorable results for and prolonged treatment are required. Currently, there is one these outcomes, the pooled effect estimates using the Hartung– large trial (38), which is recruiting 7,000 patients with MI who Knapp adjustment were not significant. However, additional have undergone PCI. The CLEAR SINERGY (Colchicine and Frontiers in Cardiovascular Medicine | www.frontiersin.org 8 June 2021 | Volume 8 | Article 676771 Diaz-Arocutipa et al. Colchicine in Acute Myocardial Infarction Spironolactone in Patients with MI/SYNERGY Stent Registry) GRADE methodology was evaluated in only one review (5). trial has a 2 × 2 factorial design where patients are randomized In contrast to prior reviews, we intended to assess the efficacy to colchicine vs. placebo and spironolactone vs. placebo with a and safety in a high risk CAD population (post–acute MI); primary outcome defined as the composite of death, recurrent therefore, the usefulness of our results is more applicable to daily target vessel MI, stroke, or ischemia-driven target vessel clinical practice. revascularization (38). Our study has some limitations. First, the number of pooled It has been largely recognized that hs-CRP is an inflammatory RCTs is low for all outcomes. However, we adjusted all 95% biomarker with important prognostic value in the setting of acute CIs using the Hartung–Knapp method to address possible type MI (39). High levels of hs-CRP following an acute coronary I error with the conventional random-effects approach. Second, event have been associated with increased short- and long-term the population of included RCTs was heterogeneous in terms risk of recurrent cardiovascular events and mortality (39, 40). of the type of patients (NSTEMI and STEMI), the dosage There are conflicting data on whether colchicine can reduce hs- of colchicine, and the timing of initiation. Third, given that CRP levels in patients with CAD (5, 8). Our study found no RCTs reported composite primary outcomes with heterogeneous significant difference on pooled hs-CRP levels during follow- definitions, they were not included in our review. Fourth, the up between patients treated with colchicine or placebo. Only treatment duration and follow-up for most of the included RCTs Deftereos et al. (19) reported a significant reduction of hs- were relatively short (≤30 days). Therefore, it is not possible CRP levels in the colchicine group. This finding may be related to know if treatment with colchicine for a longer period of to the administration of a loading dose of colchicine during time may be beneficial. Although we conducted a subgroup primary PCI. analyses according to the duration of treatment (≤30 vs. >30 Historically, colchicine has been considered a drug with days) and follow-up duration (<1 vs. ≥1 year), which showed an acceptable safety profile. The most common reported no differences between both groups, these results should be adverse effects are gastrointestinal (diarrhea, nausea, and considered with caution due to the low number of trials and vomiting), which may occur in 5–10% of patients (41). the secondary nature of the analyses. Finally, the dose and Rarely can colchicine cause myopathy, rhabdomyolysis, and duration of colchicine were inconsistent across groups. Although myelosuppression (41). Overall, our review shows the safety of subgroup analyses revealed that no dose (0.5 vs. 1 mg/day) and colchicine, with an incidence of nearly 18% of gastrointestinal duration (≤30 vs. >30 days) of colchicine reduced cardiovascular adverse events in the active group but without a pooled outcomes, these results should be considered only exploratory. significant difference when colchicine was compared with the control group (placebo or standard treatment). This finding CONCLUSION was consistent across RCTs, except for Akodad et al. (18) and Deftereos et al. (19) that reported more gastrointestinal Our meta-analysis suggests that compared with placebo or adverse events in patients treated with colchicine, which is standard treatment, colchicine does not reduce cardiovascular probably related to the use of higher doses of colchicine. mortality, recurrent MI, all-cause mortality, stroke, urgent Moreover, Tardif et al. (21) found an increase in pneumonia coronary revascularization, and levels of follow-up hs-CRP. In cases in the colchicine group in comparison to placebo addition, colchicine does not increase the risk of adverse events, (0.9 vs. 0.4%). including gastrointestinal events. However, more RCTs with There are several previously published systematic reviews larger sample sizes are needed to assess the long-term benefits or examining the effect of colchicine in patients with CAD harms of colchicine. (Supplementary Table 8) (5–11). Two reviews (5, 7) concluded that colchicine was not associated with a significant reduction of major adverse cardiovascular events or death in patients DATA AVAILABILITY STATEMENT with CAD. The other five reviews (6, 8–11) concluded that The original contributions presented in the study are included colchicine reduces the incidence of future cardiovascular events. in the article/Supplementary Material, further inquiries can be Our systematic review included RCTs that assessed colchicine in directed to the corresponding author/s. post–acute MI patients and performed further subgroup analyses according to colchicine dose, time of follow-up, and treatment duration. Conversely, prior reviews did not specify if all CAD AUTHOR CONTRIBUTIONS patients suffered from acute MI. One review included patients with ACS; however, only a narrative synthesis was conducted All authors listed have made a substantial, direct and intellectual (8). Moreover, six reviews (5–7, 9–11) pooled data from patients contribution to the work, and approved it for publication. with chronic coronary syndrome and ACS, despite the fact that these populations are clinically different. In contrast to our SUPPLEMENTARY MATERIAL findings, four previous reviews (7–9, 11) reported that colchicine was associated with an increased risk of gastrointestinal adverse The Supplementary Material for this article can be found events. Only one review used the Hartung–Knapp adjustment online at: https://www.frontiersin.org/articles/10.3389/fcvm. for CIs (5). Besides, quality of evidence per outcome using 2021.676771/full#supplementary-material Frontiers in Cardiovascular Medicine | www.frontiersin.org 9 June 2021 | Volume 8 | Article 676771 Diaz-Arocutipa et al. Colchicine in Acute Myocardial Infarction REFERENCES 20. Hennessy T, Soh L, Bowman M, Kurup R, Schultz C, Patel S, et al. The low dose colchicine after myocardial infarction (LoDoCo-MI) study: a pilot 1. Roth GA, Johnson C, Abajobir A, Abd-Allah F, Abera SF, Abyu G, et al. Global, randomized placebo controlled trial of colchicine following acute myocardial regional, and national burden of cardiovascular diseases for 10 causes, 1990 to infarction. Am Heart J. (2019) 215:62–9. doi: 10.1016/j.ahj.2019.06.003 2015. J Am Coll Cardiol. (2017) 70:1–25. doi: 10.1016/j.jacc.2017.04.052 21. Tardif JC, Kouz S, Waters DD, Bertrand OF, Diaz R, Maggioni AP, et al. 2. Libby P, Buring JE, Badimon L, Hansson GK, Deanfield J, Efficacy and safety of low-dose colchicine after myocardial infarction. N Engl Bittencourt MS, et al. Atherosclerosis. Nat Rev Dis Primers. (2019) J Med. (2019) 381:2497–505. doi: 10.1056/NEJMoa1912388 5:56. doi: 10.1038/s41572-019-0106-z 22. Wasyanto T, Yasa A, Murti B. The effect of colchicine administration on 3. Ridker PM, Everett BM, Thuren T, MacFadyen JG, Chang WH, Ballantyne C, HsCRP level and mean platelet volume in patients with miocard acute infark. et al. Antiinflammatory therapy with canakinumab for atherosclerotic disease. Indonesian J Med. (2018) 3:162–67. doi: 10.26911/theijmed.2018.03.03.06 N Engl J Med. (2017) 377:1119–31. doi: 10.1056/NEJMoa1707914 23. Tong DC, Quinn S, Nasis A, Hiew C, Roberts-Thomson P, Adams 4. Ridker PM, Everett BM, Pradhan A, MacFadyen JG, Solomon DH, Zaharris E, H, et al. Colchicine in patients with acute coronary syndrome: the et al. Low-dose methotrexate for the prevention of atherosclerotic events. N Australian COPS randomized clinical trial. Circulation. (2020) 142:1890– Engl J Med. (2019) 380:752–62. doi: 10.1056/NEJMoa1809798 900. doi: 10.1161/CIRCULATIONAHA.120.050771 5. Al-Abdouh A, Barbarawi M, Khan SU, Osman M, Upadhrasta S, Solipuram V, 24. Cronstein BN, Molad Y, Reibman J, Balakhane E, Levin RI, Weissmann et al. Colchicine therapy in patients with coronary artery disease: a systematic G. Colchicine alters the quantitative and qualitative display of selectins review and meta-analysis of randomized controlled trials. Coron Artery Dis. on endothelial cells and neutrophils. J Clin Invest. (1995) 96:994– (2020). doi: 10.1097/MCA.0000000000000931 1002. doi: 10.1172/JCI118147 6. Samuel M, Tardif JC, Bouabdallaoui N, Khairy P, Dubé MP, Blondeau L, et al. 25. Slobodnick A, Shah B, Krasnokutsky S, Pillinger MH. Update on colchicine, Colchicine for secondary prevention of cardiovascular disease: a systematic 2017. Rheumatology. (2018) 57:i4–i11. doi: 10.1093/rheumatology/kex453 review and meta-analysis of randomized controlled trials. Can J Cardiol. 26. Martínez GJ, Celermajer DS, Patel S. The NLRP3 inflammasome (2020) 37:776–85. doi: 10.1016/j.cjca.2020.10.006 and the emerging role of colchicine to inhibit atherosclerosis- 7. Ullah W, Gowda SN, Fischman D. Safety and efficacy of colchicine in associated inflammation. Atherosclerosis. (2018) 269:262– patients with coronary artery disease: a systematic review and meta-analysis. 71. doi: 10.1016/j.atherosclerosis.2017.12.027 Cardiovasc Revasc Med. (2020) 23:1–6. doi: 10.1016/j.carrev.2020.06.004 27. Tucker B, Kurup R, Barraclough J, Henriquez R, Cartland S, Arnott C, 8. McKnight AH, Katzenberger DR, Britnell SR. Colchicine in acute et al. Colchicine as a novel therapy for suppressing chemokine production coronary syndrome: a systematic review. Ann Pharmacother. (2020) in patients with an acute coronary syndrome: a pilot study. Clin Ther. (2019) 55:187–97. doi: 10.1177/1060028020942144 41:2172–81. doi: 10.1016/j.clinthera.2019.07.015 9. Tien YY, Huang HK, Shih MC, Tu YK. Drug repurposing? 28. Martínez GJ, Robertson S, Barraclough J, Xia Q, Mallat Z, Bursill C, et al. Cardiovascular effect of colchicine on patients with coronary artery Colchicine acutely suppresses local cardiac production of inflammatory disease: A systematic review and meta-analysis. J Cardiol. (2020) cytokines in patients with an acute coronary syndrome. J Am Heart Assoc. 77:576–82. doi: 10.1016/j.jjcc.2020.11.010 (2015) 4:e002128. doi: 10.1161/JAHA.115.002128 10. Xia M, Yang X, Qian C. Meta-analysis evaluating the utility of colchicine 29. Robertson S, Martínez GJ, Payet CA, Barraclough JY, Celermajer DS, Bursill C, in secondary prevention of coronary artery disease. Am J Cardiol. (2021) et al. Colchicine therapy in acute coronary syndrome patients acts on caspase- 140:33–8. doi: 10.1016/j.amjcard.2020.10.043 1 to suppress NLRP3 inflammasome monocyte activation. Clin Sci. (2016) 11. Xiang Z, Yang J, Yang J, Zhang J, Fan Z, Yang C, et al. Efficacy and 130:1237–46. doi: 10.1042/CS20160090 safety of colchicine for secondary prevention of coronary heart disease: 30. Spagnoli LG, Bonanno E, Mauriello A, Palmieri G, Partenzi A, Sangiorgi G, a systematic review and meta-analysis. Intern Emerg Med. (2021) 16:487– et al. Multicentric inflammation in epicardial coronary arteries of patients 96. doi: 10.1007/s11739-020-02606-7 dying of acute myocardial infarction. J Am Coll Cardiol. (2002) 40:1579– 12. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for 88. doi: 10.1016/S0735-1097(02)02376-8 systematic reviews and meta-analyses: the PRISMA statement. BMJ. (2009) 31. Vaidya K, Arnott C, Martínez GJ, Ng B, McCormack S, Sullivan DR, et al. 339:b2535. doi: 10.1136/bmj.b2535 Colchicine therapy and plaque stabilization in patients with acute coronary 13. Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, Welch syndrome: a ct coronary angiography study. JACC Cardiovasc Imaging. (2018) VA. Cochrane Handbook for Systematic Reviews of Interventions. 2nd ed 11:305–16. doi: 10.1016/j.jcmg.2017.08.013 Chichester: John Wiley & Sons (2019). 32. Shah B, Allen N, Harchandani B, Pillinger M, Katz S, Sedlis SP, 14. Balshem H, Helfand M, Schünemann HJ, Oxman AD, Kunz R, Brozek J, et al. et al. Effect of colchicine on platelet-platelet and platelet-leukocyte GRADE guidelines: 3. Rating the quality of evidence. J Clin Epidemiol. (2011) interactions: a pilot study in healthy subjects. Inflammation. (2016) 39:182– 64:401–6. doi: 10.1016/j.jclinepi.2010.07.015 89. doi: 10.1007/s10753-015-0237-7 15. IntHout J, Ioannidis JP, Borm GF. The Hartung-Knapp-Sidik-Jonkman 33. Bakhta O, Blanchard S, Guihot AL, Tamareille S, Mirebeau-Prunier D, Jeannin method for random effects meta-analysis is straightforward and considerably P, et al. Cardioprotective role of colchicine against inflammatory injury in a rat outperforms the standard DerSimonian-Laird method. BMC Med Res model of acute myocardial infarction. J Cardiovasc Pharmacol Ther. (2018) Methodol. (2014) 14:25. doi: 10.1186/1471-2288-14-25 23:446–55. doi: 10.1177/1074248418763611 16. Veroniki AA, Jackson D, Viechtbauer W, Bender R, Bowden J, Knapp G, 34. Fujisue K, Sugamura K, Kurokawa H, Matsubara J, Ishii M, Izumiya Y, et al. Methods to estimate the between-study variance and its uncertainty in et al. Colchicine improves survival, left ventricular remodeling, and chronic meta-analysis. Res Synth Methods. (2016) 7:55–79. doi: 10.1002/jrsm.1164 cardiac function after acute myocardial infarction. Circ J. (2017) 81:1174– 17. Wan X, Wang W, Liu J, Tong T. Estimating the sample mean and standard 82. doi: 10.1253/circj.CJ-16-0949 deviation from the sample size, median, range and/or interquartile range. 35. Forrat R, Sebbag L, Ferrera R, Hadour G, Canet E, Tabib A, et al. Effect of BMC Med Res Methodol. (2014) 14:135. doi: 10.1186/1471-2288-14-135 colchicine on circulating and myocardial neutrophils and on infarct size in 18. Akodad M, Lattuca B, Nagot N, Georgescu V, Buisson M, Cristol JP, et al. a canine model of ischemia and reperfusion. J Cardiovasc Pharmacol. (1996) COLIN trial: value of colchicine in the treatment of patients with acute 27:876–83. doi: 10.1097/00005344-199606000-00016 myocardial infarction and inflammatory response. Arch Cardiovasc Dis. 36. Bouabdallaoui N, Tardif JC, Waters DD, Pinto FJ, Maggioni AP, Diaz R, (2017) 110:395–402. doi: 10.1016/j.acvd.2016.10.004 et al. Time-to-treatment initiation of colchicine and cardiovascular outcomes 19. Deftereos S, Giannopoulos G, Angelidis C, Alexopoulos N, Filippatos after myocardial infarction in the colchicine cardiovascular outcomes trial G, Papoutsidakis N, et al. Anti-inflammatory treatment with (COLCOT). Eur Heart J. (2020) 41:4092–99. doi: 10.1093/eurheartj/ehaa659 colchicine in acute myocardial infarction: a pilot study. Circulation. 37. Shah B, Pillinger M, Zhong H, Cronstein B, Xia Y, Lorin JD, et al. Effects (2015) 132:1395–403. doi: 10.1161/CIRCULATIONAHA.115.0 of acute colchicine administration prior to percutaneous coronary 17611 intervention: COLCHICINE-PCI randomized trial. Circ Cardiovasc Frontiers in Cardiovascular Medicine | www.frontiersin.org 10 June 2021 | Volume 8 | Article 676771 Diaz-Arocutipa et al. Colchicine in Acute Myocardial Infarction Interv. (2020) 13:e008717. doi: 10.1161/CIRCINTERVENTIONS.119.0 Conflict of Interest: VP was employed by the company MedErgy HealthGroup, 08717 Inc. 38. Siniorakis E, Arvanitakis S, Nikolopoulos I, Elkouris M. COVID-19 interference with renin-angiotensin system in the context of heart failure. The remaining authors declare that the research was conducted in the absence of mBio. (2020) 11:e00946–20. doi: 10.1128/mBio.00946-20 any commercial or financial relationships that could be construed as a potential 39. Eggers KM, Lindahl B. Prognostic biomarkers in acute coronary syndromes: conflict of interest. risk stratification beyond cardiac troponins. Curr Cardiol Rep. (2017) 19:29. doi: 10.1007/s11886-017-0840-3 Copyright © 2021 Diaz-Arocutipa, Benites-Meza, Chambergo-Michilot, Barboza, 40. He LP, Tang XY, Ling WH, Chen WQ, Chen YM. Early C-reactive Pasupuleti, Bueno, Sambola and Hernandez. This is an open-access article protein in the prediction of long-term outcomes after acute coronary distributed under the terms of the Creative Commons Attribution License (CC BY). syndromes: a meta-analysis of longitudinal studies. Heart. (2010) 96:339– The use, distribution or reproduction in other forums is permitted, provided the 46. doi: 10.1136/hrt.2009.174912 original author(s) and the copyright owner(s) are credited and that the original 41. Slobodnick A, Shah B, Pillinger MH, Krasnokutsky S. Colchicine: old publication in this journal is cited, in accordance with accepted academic practice. and new. Am J Med. (2015) 128:461–70. doi: 10.1016/j.amjmed.2014. No use, distribution or reproduction is permitted which does not comply with these 12.010 terms. Frontiers in Cardiovascular Medicine | www.frontiersin.org 11 June 2021 | Volume 8 | Article 676771