1. Identifying the research questions

Consistent with the standard recommendations for scoping reviews, we convened an initial meeting of the core review team members and identified the broad primary research questions: 1) What is the involvement of the cardiac system in adult patients diagnosed with COVID-19 and, 2) What are the interactions between SARS-CoV-2 infection and ACEI and ARB treatment, 3) What are the main pathophysiologic mechanisms of cardiac manifestations in Covid-19.

2. Identifying relevant studies

A search strategy was developed by an experienced author (LR) and a librarian (GR) (S1 Table). The search was conducted across Embase, PubMed, Epistomonikos, MedRxiv, BioRxiv databases, from its inception to November 2020; no language restrictions were applied. We manually searched the reference lists of systematic reviews that were included, as well as including relevant studies recommended by experts. Additional searches were performed to ensure that new existing systematic reviews did not importantly modify key findings.

3. Study selection

Explicit a priori eligibility criteria (S2 File) were applied at title and abstract, and full text screening. The principal inclusion criterion was a systematic review that addressed one or more of the study objectives. Primary peer-reviewed articles and nonpeer -reviewed articles were considered relevant if they addressed one or more aspects of the research question. One reviewer uploaded the literature search results into Microsoft excel®, and screened the titles and abstracts of all references, and full texts of the identified studies. Included studies were examined by two other reviewers to confirm inclusion, extract relevant data, and map them to specific research questions. Exclusion criteria were also recorded.

Inclusion criteria: Systematic reviews published from inception to November 2020, on diagnosed COVID-19 adult patients, without restrictions on race, gender, geographical location or setting, reporting: cardiac symptoms/complications; cardiac biomarkers, imaging, clinical management for cardiac complications; cardiac adverse events in those on COVID-19 therapeutics; adverse/improved general and cardiac outcomes in patients on RAASI; and pathophysiology of the cardiovascular system involvement in SAR-CoV-2.

Exclusion criteria: Systematic reviews reporting on Kawasaki-like syndromes, multi-system inflammatory syndrome (MISC) related to COVID-19, studies enrolling pediatric samples; studies comparing Covid-19 with previous SARS or Middle East Respiratory Syndrome (MERS) infections; animal studies, in vitro experiments, drug modelling, and other unrelated aspects of COVID-19 research.

4. Charting the data

A data extraction form (MS excel) was developed and validated by two reviewers (SP and LR). For each study, we extracted data on source, patient demographic, clinical topic covered and outcomes (S2 File).

5. Collating, summarizing, and reporting the results

The key information from each article, as described in the data items section, are reported as a narrative and using descriptive statistics using tables and graphs, with no attempt to aggregate findings from various studies, as this is not the aim of a scoping review. We used the qualitative characteristics and the numerical distribution of mapped evidence addressing each of the research questions to formulate our recommendations for future research.

Risk of Bias assessment: Two reviewers (SP and VS) assessed the risk of bias across eligible systematic reviews using the tool ROBIS [19]. Any discrepancy was resolved by a third reviewer (LR).

Characteristics of included studies

Table 1 summarizes the main characteristics of the 63 systematic reviews (S2 Table includes study title, author, and reference number) that include 46 (73%) peer-reviewed articles, and 17 (27%) non-peer reviewed studies. Fourteen (22.2%) studies were systematic reviews and 49 (77.8%) were systematic reviews with meta-analysis. Fifty-six systematic reviews reported on the origin of the primary studies (n = 1575), most of which come from upper middle-income or high-income countries: China 938 (59.6%), USA 202 (12.8%), or Italy 110 (7.0%) (Figure 2 and S3 Table). Eleven studies reported data from a single country, which was from China. Fifty studies reported total patient population, ranging from 12 to 2,065,805 patients. Forty-five (90%) of the 50 reporting studies had more than 1,000 patients each, and 15 (30%) studies reported on more than 10,000 patients per study. The research domains of the included studies were acute cardiac injury, arrhythmias, other CV complications (heart failure, cardiogenic shock, cardiomyopathy, acute coronary syndrome), cardiotoxic therapeutics, RAASI use and pathophysiology of cardiac involvement in COVID-19 patients.

The greatest number of cardiac studies are shown to have been published in October (S1 Figure) The highest number of primary cardiac studies of COVID-19 patients included in the systematic reviews throughout the year 2020 is from China (S1 Figure).

From the 33 systematic reviews that reported the gender, 30 (91%) studies had predominant COVID-19 male population (>50%). From the 37 studies reporting the mean or median age of the patients, 25 (67.6%) included patients that were 40 years and above, and 36 (97.2%) included patients that were 20 years and older.

Risk of bias assessment: Of the 63 studies we evaluated, 45 studies had a low risk of bias, while nine studies had a high risk of bias and nine were unclear (S2 Figure).

1) What is the involvement of the cardiac system, in adult patients diagnosed with COVID-19

The reported cardiac manifestations in COVID-19 patients are summarized in Tables 2, 3, and 4 (S4 Table and S3 Figure).

The most frequent reported cardiac complications were arrhythmias (S3 Figure and S4 Table), myocarditis (Table 5), heart failure, cardiogenic shock, cardiac arrest, and acute coronary syndromes/CAD. One study (n = 12) reporting on 12 patients with Takotsubo syndrome reported acute cardiac injury defined as elevated troponin I, in 75% of the cases [20].

Incidence of arrythmias ranged from 1.8% to 26% in hospitalized COVID-19 patients [21, 22, 23, 24, 25, 26, 27, 28, 29], while an incidence of 33% to 48% was reported in severe disease [21, 25, 26, 29, 30]. Newly occurring arrhythmias in COVID-19 patients predicted severe disease and/or mortality [21, 25, 30, 31, 32, 33, 34]. The most reported arrhythmias were ventricular arrhythmias (ventricular tachycardia, ventricular fibrillation) [24, 29], and atrial arrhythmias (atrial fibrillation, atrial flutter) [24, 27, 29].

Ten studies aimed to report on the cardiac adverse events due to pharmacological treatment for COVID-19 and gave inconsistent results [35, 36, 37, 38, 39, 40, 41, 42, 43, 44]. While three studies reported a significant association between arrythmias and HCQ or AZT [424344], four studies did not [36, 37, 38, 41]. Two studies suggest a probable cardioprotective role of HCQ [36, 37] in COVID-19 patients.

Drug induced long QT syndrome (DILQ) was reported with AZT and HCQ/CQ [29, 40, 44], and monotherapy with COVID-19 repurposed medicines [35, 41, 42, 43, 44], while other studies did not [39, 40, 44].

COVID-19 patients with severe disease and non-survivors had more prominent laboratory abnormalities (S5 Table), specifically cardiac biomarkers including increased levels of Troponin I [23, 24, 30, 31, 32, 33, 45, 46, 47, 48, 49, 50, 51]. Elevated levels of creatine kinase MB isoenzyme (CK-MB) [23, 24, 30, 31, 33, 34, 48, 49, 50, 51, 52] NT-BNP [24, 30, 31, 32, 33, 34, 48, 49, 50, 51, 53, 54, 55], lactate dehydrogenase (LDH) [23, 32, 34, 46, 51] and D-dimer levels [23, 50, 51, 54] were also reported and associated with disease severity and mortality.

Four studies focused on myocarditis [24, 26, 48, 49]. The general clinical symptoms observed in COVID-19 patients with myocarditis are similar to those observed in a respiratory tract infection such as dyspnea, chest pain, fever, and cough. These patients showed ST elevation (25%–55.6%) and T-wave inversion (25%–33.3%) on ECG, reduced left ventricular ejection fraction (LVEF) (50%–66.7%) on ECHO, and diffuse/late gadolinium enhancement on cardiac MRI (100%). Shafi et al. reported the use of CMRI with gadolinium washout, combined with ECHO findings to confirm myocarditis [24]. Glucocorticoids, immunoglobulins, and antivirals were the most used pharmacotherapy.

2) What are the interactions between SARS-CoV-2 infection and ACEI and ARB treatment

Eight (66.7%) out of 12 systematic reviews found no association between the use of ACEI or ARBs and COVID-19 outcomes (Table 6 and S6 Table) such as susceptibility to infection, hospitalization, severity and mortality, while some studies found no association even after adjusting for potential confounding factors [67, 68, 69, 70, 71, 72, 73, 74]. However, in one study the use of RAASI was found to decrease the length of hospital stay [67]. Other studies including a study in which data was collected on 2,065,805 individuals [75] suggest a protective role of RAASI, decreasing the risk of severe illness and mortality [74, 75, 76, 77, 78, 79, 80].

Six of the seven studies reporting subgroup analyses reported no difference in outcomes based on the type of RAASI [67, 71, 72, 74, 75, 78], while only one study reported that taking ACEIs were better in decreasing the severity and mortality compared with those receiving ARBs [77].

3. What are the main pathophysiologic mechanisms of cardiac manifestations in Covid-19?

Five systematic reviews reported on the cardiac pathology in COVID-19 patients (S7 Table) [45, 81,8283, 84]. The reported mechanisms of myocardial injury considered plausible were: hyperinflammation and cytokine storm, mediated through pathologic T cells and monocytes, leading to myocarditis; respiratory failure and hypoxemia resulting in damage to cardiac myocytes; downregulation of ACE2 expression and subsequent protective signaling pathways in cardiac myocytes; hypercoagulability and development of coronary microvascular thrombosis; diffuse endothelial injury; and inflammation and/or stress causing coronary plaque rupture or supply-demand mismatch leading to myocardial ischemia/infarction (MI).

Clinical presentation/feature

The most prevalent cardiac presentation in SARS-CoV-2 infection are arrhythmias, heart failure and acute coronary syndromes. These presentations were similar across the global regions; however, the origin of the primary studies was dominated by three countries, China, the USA, and Italy, that were epicenters for Asia, the Americas, and Europe, respectively.

Cardiac markers

The most widely cited laboratory indicator of acute cardiac injury in COVID-19 is elevated serum levels of troponin I, which is released from damaged cardiomyocytes [85]. Usually, acute cardiac injury was defined as cTn >99th percentile. This definition may be questioned since plasma levels of ultra-sensitive troponins are frequently increased in many settings such as respiratory failure in the absence of acute coronary syndromes or myocarditis [86]. Other cardiac biomarkers, CKMB, BNP and LDH were associated with COVID-19 disease severity and cardiac injury. How these cardiac biomarkers may be used for treatment adaptation needs to be evaluated. An increase in D-dimer levels in these patients may be due to a thrombogenic state caused by the elevated proinflammatory cytokines, suppressing cardiac function directly, damaging endothelial cells and amplifying vascular inflammation [87]. Little is said in these reviews about the role of cardiac infection in the myocardial injury. Recent pathological findings suggest that the SARS-CoV-2 infection of the heart is very uncommon, even in the severe forms [88].

Myocarditis

It is difficult to estimate true prevalence from the available literature, given that three of the four papers [24, 48, 49] were case series only (100% prevalence, by definition, because only positive cases are included). The remaining study [26] that included 101 patients reported a prevalence of 19–28%; however, this is likely an overestimate of true prevalence given that it is a selected cohort, and few additional details of the patients were provided. The first cohort of myocarditis patients was reported from China [89], with subsequent case reports and case series from USA and Europe [24, 48, 49]. Presenting symptoms include fatigue, shortness of breath, and chest pain. Tachyarrhythmias are a potential cause of clinical deterioration, but overall prevalence of this complication remains unclear. Clinical diagnosis is made in by looking for typical myocarditis characteristics (symptoms plus elevated troponin and B-type natriuretic peptide) and myocardial ischemia and cardiomyopathy should also be considered in the differential [90]. Notably, electrocardiographic changes seen in pericarditis (widespread ST segment elevation, t-wave inversion, PR segment depression) may not been seen in myocarditis [91]. Three studies in our review reports the use of echocardiography and cardiac magnetic resonance imaging (CMRI) for the diagnosis of myocarditis [24, 48, 49], in line with the American Heart Association, which states the requirement of endomyocardial biopsy as a definitive diagnosis and recommends a contrast CT in the absence of a CMRI [92]. No specific, COVID-19-focused myocarditis therapies exist; therefore, patients are largely managed supportively.

Arrhythmias

Consistently, all the systematic reviews found that cardiac arrhythmias were significantly associated with an increased risk of a poor outcome in COVID-19 patients. COVID-19 patients are prone to the development of arrhythmias, especially supraventricular and ventricular tachycardia, indicative of myocardial injury and or hemodynamic instability and often observed in severe or critical situations. This is line with that was described in patients with severe acute respiratory failure from other origins. This suggests the usefulness of appropriate monitoring by electrocardiogram (ECG) in severe forms.

COVID-19 therapeutics and arrhythmia/LQTS

Since the outbreak of COVID-19, several medicines have been proposed and are being evaluated in intensive clinical trials in COVID-19 patients. HCQ and CQ are traditional antimalarial and autoimmune disease drugs that have been shown to control the SARS-CoV-2 infection in vitro [93], although the cardiotoxicity of HCQ and CQ should not be neglected. Several systematic reviews included in our study showed an association between these drugs and an increase in serious ventricular arrhythmia [21, 22, 29]. Since these reviews have been published, several randomized controlled trials have been performed. A recent Cochrane review of these randomized trials [94] concluded that adverse events are tripled in patients on hydroxychloroquine or chloroquine compared to placebo, but very few serious adverse events were found. No significant increase of torsade de pointe incidence was found on hydroxychloroquine.

Furthermore, the combination of HCQ/CQ and azithromycin was shown to be associated with an increased in QT prolongation incidence and fatal cardiac complications in a cardiac-impaired population [39, 40, 44, 95] but not significant with HCQ/CQ monotherapy. Some of the studies reporting LQTS have not defined the methodology used for QTc prolongation. Concomitant medications, electrolyte disturbances, structural heart disease and advancing age may also prolong the QT interval in COVID-19 patients, therefore, regular monitoring the QT interval in hospitalized patients or with the use of biodevices for outpatients maybe helpful in this cohort of patients.

Many of the clinically important therapeutic drug classes for the management of COVID-19 patients such as antihistamines [96], beta-agonist [97], analgesics, antipsychotics [98], are known to cause cardiotoxicity. Given the pivotal role of the immunologic overresponse in COVID-19, anti-inflammatory therapy has been used for treating COVID-19 as well as some of the newer medications such as monoclonal antibodies, tocilizumab, interferon [99]. Unfortunately, our search did not capture systematic reviews reporting on cardiac events related to these medications.

RAASI

Most of the reviews included in our work did not find any relation between RAASI and COVID-19 outcome. Some suggest a protective effect. Since the publication of these systematic reviews, several RCTs have addressed this question [100, 101]. In the largest study, R Lopes et al. [101] found that there was no significant difference in the mean number of days alive and out of the hospital for the patients assigned to discontinue vs continue RAAS inhibitors in a RCT including among 659 patients hospitalized with mild to moderate COVID-19. Given the common use of RAASIs worldwide, based on the available evidence and expert consensus, RAASI should not be systematically interrupted in patients with COVID-19.

Clinical implications and Research Gaps

Our findings indicate that COVID-19-related cardiac manifestations are common and, in many cases, have clinically important consequences. In terms of current clinical management and future research priorities, our review suggests the following:

1. Clinicians should assess for cardiac disease as clinically indicated. However, whether diagnostic management should change to detect cardiac disease and if this improves outcomes remains unknown.
2. Contemporary guideline recommendations intended for non-COVID-19 patients with cardiac disease should also be applied to the treatment of cardiac disease in patients with COVID-19 unless specific evidence exists to suggest otherwise. This is because current knowledge is at an early stage and there are no targeted therapies available for specific COVID-19 related cardiac complication. Research is needed to verify that this is the best approach and identify focused therapies. In terms of research priorities, it seems prudent to align these with the severity of the potential complication, with acute cardiac injury, arrhythmia, and myocarditis being the most serious. Since recognition of these conditions is comparatively well established (both in patients with COVID-19 and in those without), identification and testing of specific therapies for COVID-19 in general, and cardiac complications, should be prioritized.
   1. Given the complex interplay of SARS-CoV-2 with the cardiac system, randomized controlled trials are urgently needed to investigate treatment modalities to reduce the incidence and complications associated with COVID-19 related acute myocardial injury.
   2. Despite detection of SARS-CoV-2 RNA/viral particles in myocardial tissue, no clear diagnosis of myocarditis could be made histopathologically [49]. Improved understanding of the pathophysiology of COVID-19-related complications such as myocarditis should be a priority for future research.
   3. Potential grants/funding directed to better understand the inter-relationship between cardiac disease and COVID-19, addressing both research platforms and integrated multidisciplinary health systems that deliver care at the same site.
   4. Data on the long-term consequences of COVID-19 in the heart have not been studied in the systematic reviews in our scope. There is a need for ongoing investigation of the long-term cardiac consequences of COVID-19, regardless of previous (cardiovascular) diseases or the severity of the disease.
   5. There is many anecdotal reports and some studies on the cardiac complications following vaccination for COVID-19. There is a need to synthesize the available data regarding cardiac symptoms in post-vaccinated population.
3. However, until this research is done, patients should not be denied important, evidence-based therapies (e.g., RAASI for hypertension and heart failure) based on speculation.
4. Optimal pathways of care for patients with COVID-19 with cardiac complications need to be identified. Often, important testing is delayed because of isolation requirements and/or concerns regarding transmission risk. Balancing the need to optimize care versus health care provider safety is required.
5. Although is often tempting to initiate unproven therapies in severe cases of COVID-19, clinicians should be alerted to the importance of avoiding non-evidence-based treatments because potential harm may result.

Limitation

Due to the urgent need for evidence on this topic and limited time, we did not contact authors to clarify the details of primary data when necessary. Studies on this topic are rapidly being conducted and may not have been included in this review when results were published. Many studies have shown the significant association of cardiovascular comorbidity with COVID-19 severity and mortality, therefore was not examined in this review. Given the lack of data in some of the studies, we were unable to gather information on the percentage of COVID-19 inpatients and ambulatory patients. In addition, publication bias may have affected the estimates reported here, as nine studies were assessed as high risk of bias. Limiting the scope to systematic reviews did not allow to consider recent high-quality randomized studies that could address therapeutic questions.