Introduction

Chagas disease, also known as American trypanosomiasis [1], is an important public health problem in Latin America, and remains a substantial cause of morbidity and mortality [2]. Recently, it has become a health concern [1] in areas such as Europe, North America, Japan and Australia, especially due to immigration from endemic areas to developed countries [3,4]. About 6 to 8 million people are affected worldwide [5], and less than 1% of infected people have access to adequate diagnosis and treatment [6].

The clinical course of the disease is extremely variable, and although many individuals remain asymptomatic for long periods, approximately 30% of infected people develop Chagas cardiomyopathy [7]. Cardiac complications result in remodeling of the cardiac collagenous matrix and subsequent fibrosis, leading to increased myocardial stiffness, systolic and diastolic dysfunction, and ultimately a severe dilated cardiomyopathy associated with ventricular arrhythmias [8].

Since trypanocidal therapy with benznidazole in patients with established Chagas cardiomyopathy had not proved to reduce cardiac clinical deterioration in clinical trials [9], the treatment of heart failure and arrhythmias became the main therapeutic strategy to curb the evolution of the disease. Despite the lack of solid evidence-based data on Chagas disease, therapy should generally be instituted extrapolating from guidelines for the management of heart failure patients, including neurohumoral inhibition, resynchronization therapy, and implantable cardioverter defibrillator [10]. However, this alternative is not readily available for low income populations such as those affected by Chagas disease. Studies have shown that amiodarone can improve survival in patients who have a high risk of arrhythmic death [11,12]; thus, amiodarone has been recommended as the treatment of choice for all patients with sustained ventricular tachycardia, and for those with no sustained ventricular tachycardia and myocardial dysfunction [13]. In a recent meta-analysis of primary prevention (17 studies, 8383 participants), amiodarone reduced sudden death, cardiac mortality and all-cause mortality; however, Chagas patients were not included. Nevertheless, amiodarone was associated with increased adverse effects, both hypo- or hyperthyroidism and pulmonary fibrosis, and increased risk of discontinuation compared with placebo [14].

Due to the lack of direct evidence regarding the benefits and risks of amiodarone use in Chagas patients, evidence from patients with other cardiopathies is often used and extrapolated to patients with Chagas disease for decision-making. However, there are some particularities of the amiodarone effect on Chagas disease; for instance, there is some evidence that amiodarone also has anti-T. cruzi activity, disrupting Ca2+ homeostasis and blocking oxidosqualene cyclase in T. cruzi, causing ultrastructural damage [15]. Only small studies have assessed the effects of amiodarone specifically in Chagas disease, mostly published in Portuguese and Spanish, and it is not clear if the estimated effect is similar to that in the general population. Therefore, our aim was to do a systematic review of the effect of amiodarone in patients with Chagas disease, to provide a more accurate estimate of the effect of amiodarone in this population and to identify the potential benefits and harms of this drug.

Methods

Results

Description of studies

Nine studies [26–34] from 417 records found in the database search fulfilled eligibility criteria and were included in this systematic review, providing data from 365 subjects. There were three before-after studies [30,32,34], five case series (1 of which as follow period of a cluster randomized trial) [26,27,29,31,33] and one randomized clinical trial [28]. The mean age of participants ranged across the trials from 15 to 78 years. In 8 studies [26–33], amiodarone was administered orally and the dose ranged from 200 to 1200 mg, with a mean treatment time ranging from 1 day to 27 months. In 1 study [34], the dose of amiodarone was 5 mg/kg, given intravenously.

Fig 1 shows the flow diagram of the selection process, and Table 1 summarizes the characteristics of the included studies.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Flowchart of included studies.

https://doi.org/10.1371/journal.pntd.0006742.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Characteristics of studies included in systematic review.

https://doi.org/10.1371/journal.pntd.0006742.t001

For data synthesis, only 2 studies [30,32] were included in the IPD meta-analysis for ventricular premature beats, ventricular tachycardia and ventricular couplets. Other studies [26–29,31,33,34] had different designs or presented the results in a non-comparable manner; therefore, these results were described narratively (Table 2).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Narrative results.

https://doi.org/10.1371/journal.pntd.0006742.t002

Effects of interventions

Ventricular tachycardia.

Seven studies [27,28,30–34] evaluated ventricular tachycardia, and two studies [30,32] were included in the IPD meta-analysis. In these two studies, the median number of ventricular tachycardia episodes in 24-hour Holter was 25 (IQR 3–856) in the control group and 0 (IQR 0–0) for the intervention group (Table 3). Amiodarone reduced ventricular tachycardia in 99.9% (95%CI 99.8–100% in 24-hour Holter monitoring) (Table 3). All studies that were not included in the meta-analysis reported a decrease in the number of ventriculartachycardia episodes after the use of amiodarone (Table 2). The quality of evidence according to GRADE for this outcome was moderate. (S4 Appendix).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Ventricular tachycardia, ventricular premature beats and ventricular couplets after amiodarone.

https://doi.org/10.1371/journal.pntd.0006742.t003

Ventricular premature beats.

Eight studies [26,28–34] evaluated ventricular premature beats and two studies [30,32] were included in the IPD meta-analysis. In these two studies, the median number of ventricular premature beats in 24-hour Holter was 8924.5 (IQR 5987.5–21295.5) in the control group and 161.5 (IQR 34.3–560.3) for the intervention group (Table 3). Amiodarone reduced ventricular premature beats in 93.1% (95%CI 82–97.4% in 24-hour Holter monitoring) (Table 3). All studies that were not included in the meta-analysis reported a decrease in the number of ventricular premature beats after the use of amiodarone (Table 2). The quality of evidence according to GRADE for this outcome was moderate (S4 appendix).

Ventricular couplets.

Three studies [28,30,32] assessed ventricular couplets and two of them [30,32] were included in the IPD meta-analysis. In the control group, 38 out of 38 patients had ventricular couplets (100%), while 8 out of 38 patients (21.0%) had it in the intervention group (Table 3). Amiodarone reduced the incidence of ventricular couplets by 79.0% (RR 0.21, 95%CI 0.11–0.39 in 24-hour Holter monitoring) (Table 3). In the study that was not included in the meta-analysis, 95.2% (38/40) of patients showed a reduction in the number of ventricular couplets with use of amiodarone [28]. The quality of evidence according to GRADE for this outcome was moderate (S4 appendix).

Sudden death.

Three studies [27,31,32] (n = 107) evaluated sudden death after the use of amiodarone. Eight patients (7.50%) experienced sudden death after 5 to 27 months of treatment. The quality of evidence according to GRADE for this outcome was very low (S4 appendix).

Side effects.

All studies [26–34] investigated side effects caused by amiodarone and 7 were included in the meta-analysis [27,28,30–34]. The studies not included in the meta-analysis are described in Table 2. Five studies [27,28,30–32] assessed corneal microdeposits, three studies [30–32] reported cases of gastrointestinal events, six studies [27,30–34] reported cases of sinus bradycardia, three studies assessed dermatological events [27,28,32] and five studies assessed drug discontinuation [27,28,31–33].

The prevalence of corneal microdeposits after use of amiodarone was 61.0% (95%CI 18.8–91.3%, n = 161), gastrointestinal events 16.1% (95%CI 6.61–34.2%, n = 47), sinus bradycardia 12.6% (95%CI 3.68–35.4%, n = 144), dermatological events 10.5% (95%CI 4.87–21.3%, n = 99), and drug discontinuation 7.59% (95%CI 4.12–13.6%, n = 156) (S5 Appendix). Three studies assessed hypo- or hyperthyroidism (n = 203), where only one patient showed hypothyroidism [27,29,31]. One study reported two cases of pneumonitis during the use of amiodarone [27].

Discussion

This systematic review included 9 studies that determined the effects of amiodarone with regard to arrhythmia, side effects and sudden death in individuals with Chagas disease. Due to the high heterogeneity on outcome assessment and diversity of study designs, most results were presented qualitatively. However, it was possible to pool two before-after studies in an IPD meta-analysis, since these studies provided the full dataset in the published articles [30,32]. For adverse side effects, available data allowed pooling frequency estimates using the conventional meta-analysis approach for seven studies [27,28,30–34].

According to our findings, there was moderate certainty in the body of evidence for the effect of amiodarone on the clinically significant reduction of arrhythmia in patients with Chagas disease, such as ventricular tachycardia episodes (reduction of 99%), ventricular premature beats (reduction of 93%) and ventricular couplets (RR = 0.21). Although it was only possible to pool two studies in the IPD metanalysis, all other studies showed similar effects on these outcomes.

Despite the strong evidence for arrhythmia reduction, studies did not provide enough information for the assessment of the effect on clinically relevant outcomes, such as mortality and hospitalization. Arrhythmia was mainly assessed using 24-hour Holter or conventional electrocardiogram evaluation, and it was expected that most episodes would be asymptomatic, without any clinical impact. Although clinical benefit is expected with arrhythmia reduction, this information should be interpreted with caution, since arrhythmia is a surrogate outcome, and is not clear the clinical impact on death and hospitalization reduction over time. Little information was identified related to hard endpoints; three studies, with 107 patients, observed 8 sudden deaths (7.50%) after 5 to 27 months of treatment, without comparison with any control group.

Effects of amiodarone on sudden death are available from different clinical conditions. Piccini et al. performed a meta-analysis of studies using amiodarone for primary prophylaxis of sudden cardiac death in high-risk patients (post myocardial infarction, heart failure). Compared to placebo/control, there was a 29 and 18% reduction in sudden death and cardiovascular mortality, respectively [35]. Moreover, in a recent meta-analysis of primary prevention, amiodarone reduced the risk of sudden cardiac death by 34% [14]. In the absence of direct evidence for patients with Chagas cardiomyopathy, these figures may provide an indirect estimate of the potential benefit in our population of interest.

Regarding side effects, our systematic review observed that amiodarone was associated with corneal microdeposits, gastrointestinal events, sinus bradycardia, dermatological events, pneumonitis, hypothyroidism and drug discontinuation. Other studies that did not include Chagas patients suggested that adverse effects are common, with a prevalence as high as 15% in the first year of use and 50% during long-term use, indicating a cumulative effect with chronic use, corroborating our results [36]. Moreover, in an additional study, the risk of discontinuation of amiodarone was 45% superior than placebo, however presenting similar rates when compared with other antiarrhythmics [14]. Of note, even though our review found that amiodarone was associated with a low incidence of sinus bradycardia (7%), concomitant use of amiodarone and beta-blockers in Chagas cardiomyopathy may increase the risk of exacerbation of bradyarrhythmia [37].

There are other interventions with potential benefit for patients with Chagas disease and arrhythmias. Observational studies have shown that an implantable cardioverter defibrillator is effective for primary or secondary prevention of death in Chagas disease [38,39]. However, there are obstacles concerning its use due to socioeconomic limitations such as low income and poor health and nutrition conditions. The first randomized clinical trial comparing the implantable cardioverter defibrillator and amiodarone for primary prevention of death in patients with Chagas disease and nonsustained ventricular tachycardia, the CHAGASICS study, is ongoing and may provide important information for the field in the near future [40].

Our systematic review and meta-analysis has several strengths. We performed IPD meta-analysis, which has potential advantages such as increased power and reduced bias, and it is considered the gold standard in evidence synthesis [41]. Moreover, our comprehensive search of various databases, along with seeking the information from study authors and experts in the field, makes it unlikely that any relevant study was missed.

However, our review has some limitations, especially related to the nature of the available data. Most of our results are based on limited observational data. All included studies were small and presented methodological limitations, and the majority of them presented a short-term follow-up.

Another fact that may reduce the external validity of our findings is that all included studies were published from 1980 to 1990. In the last 30 years, the treatment of heart failure and arrhythmias has progressed exponentially. The addition of ACE inhibitors, beta-blockers and mineralocorticoid antagonists has modified the evolution of heart failure, and the emergence of electrophysiology as well has added the use of ablation therapies and implantable cardio-defibrillators in the management of patients with Chagas cardiomyopathy. There is a gap in the literature showing the effect of such developments on the number of events or mortality in these patients.

An important aspect to highlight is the lack of interest of the scientific community regarding this topic, where the most recent article included in the review was published in 1990, and the absence of systematic reviews targeting this research question. Our study demonstrates the lack of clinical trials and observational data on therapeutic options for chagasic cardiomyopathy considering the current therapeutic arsenal.

In spite of data limitations, the estimates reported here show that amiodarone seems to be an effective antiarrhythmic drug for patients with Chagas disease, reducing uncomfortable symptoms such as palpitations and potentially fatal events related to ventricular tachycardia. However, it is important to stress that there is no direct evidence for the reduction on clinically relevant outcomes. Our findings may be useful to support clinicians’ decision-making in amiodarone use, especially in settings where an implantable cardioverter defibrillator is not available or affordable, but there is a need to balance the potential benefits and harms at the individual level.

Supporting information