Cardiac dysfunction in Multisystem Inflammatory Syndrome in Children: An Italian single-center study

Multisystem inflammatory syndrome in children (MIS-C), also called pediatric inflammatory multisystem syndrome, is temporally related to SARS-CoV2 infection. It is a rare but severe hyperinflammatory disease that affects pediatric patients, typically 3–6 weeks after SARS-CoV2 contact [1].

While children are largely spared from severe acute SARS-CoV2 infection, probably due to under-expression of ACE2 receptors [2], MIS-C is a condition with clinical features similar to incomplete Kawasaki disease, macrophage activating syndrome, toxic shock syndrome, and septic shock [3, 4]. It affects multiple organs, the cardiac, gastrointestinal, muco-cutaneous, respiratory, neurological, and hematological systems [5, 6].

MIS-C was first described in April 2020 in Europe, but nowadays reports and papers exist from all over the world. The incidence is generally unknown, but a recent US paper reported an adjusted estimated incidence from 1 to 10 cases per 1.000.000 people per month [7]. Male patients and Hispanic and African people have a higher prevalence [8, 9].

The majority of patients are previously healthy children, without relevant comorbidities, with the exception of asthma and obesity [5].

The pathophysiological mechanism of the disease is unclear. MIS-C cases have a 3–6-week lag from the peak of cases in adults. The proposed mechanism is a post-infectious phenomenon mediated by IgG antibodies. Apparently, acute infection of SARS-CoV-2 induces a proinflammatory reaction and, in a genetically susceptible child, a delayed hyperinflammatory reaction of vasculitis with augmented levels of lymphocyte T-helper 17 and T-helper 1 mediators, and a cytokine storm, including massive release of inflammatory mediators and exaggerated activation of the immune system [10]. Chang et al., in a recent study, supported the role of a skewed cytokine profile towards IL6/IL8 pathways in causing heart dysfunction [11].

The clinical manifestations include persistent fever and involvement of at least two organ systems. Gastrointestinal symptoms, including vomiting, diarrhea, and abdominal pain, are reported in the vast majority of cases, from 50 to 100%. Rash, mucositis, conjunctivitis, and peripheral edema are frequently observed, in 36–81% of cases. Neurological symptoms are usually mild (headache, confusion, and irritability) and occur in 12–56% of cases, but rarely can be more severe, including meningitis and seizures. Respiratory symptoms are not a significant part of the MIS-C presentation, but shortness of breath, cough, hypoxia, and respiratory distress have been documented in up to 65% of cases [9, 12, 13].

Cardiac system involvement is frequently described and echocardiography, electrocardiography, and even laboratory abnormalities included. Acute myocardial dysfunction is the most prevalent, with generally mild or moderate depressed ejection fraction (EF). Only one third of patients present with severely reduced EF (EF < 30%). Cardiogenic or vasodilatory shock are described in up to a half of MIS-C diagnoses, and require inotropic support, while 28% need mechanical circulatory support [5].

Coronary arteries dilation and aneurysms are frequently described, but the prevalence varies significantly among reports, and it is not always easy to understand the rate of coronary enlargement and involvement. The most recent cardiological review reports an incidence of 8–24% in the larger studies. Most of the coronary involvement is mild (dilation or small aneurisms, Z score 2.5–5) and transient, but there have been a few reports of large aneurysms (Z score > 10) [14].

The aim of this study was to describe cardiac involvement in patients with MIS-C, defining the role of severe cardiac dysfunction in determining the clinical presentation and outcomes in a single cohort of pediatric patients.

The recent pandemic of COVID-19 was characterized by poor clinical involvement of children compared to adults, but a novel MIS-C, mediated by cytokine activation, has been reported. This syndrome arises 3–5 weeks after no evidence of active current COVID-19 infection, and causes systemic multi-organ inflammation. In mid-May 2020, the CDC published a case definition for MIS-C for disease surveillance [15]. The heart is one of the main organs involved, and cardiovascular complications triggered recommendations for immunomodulatory treatments. Ventricular dysfunction, pericardial effusion, and valvulitis are commonly diagnosed in MIS-C.

We reported cardiac involvement in 81% of our patients, confirming that the heart, together with the gastrointestinal system, represents the most common target organ in patients with MIS-C. Only 28% of children showed a preserved EF, 41% patients had a mildly reduced EF, and 16 and 16% exhibited moderately and severely depressed functioning, respectively. A recent review of cardiac involvement in patients with MIS-C reported left ventricular dysfunction in 31 to 58% of patients [14]; similarly, in a recent study, Leora R. Feldstein et al. reported 34.2% of patients had a depressed LVEF [12].

The age distribution of our population supports a greater involvement of older children and adolescents, (median age of 10 years), as already reported in the literature [9, 22].

Fortunately, in accordance with other previous studies [23‐25], we observed a prompt response to therapy (3 ± 2 days after treatment). At discharge, all patients had an LVEF > 55%.

It is known that most patients recover within days to weeks and mortality is rare, although the medium- and long-term sequelae, and particularly cardiovascular complications, are not yet known [14, 22, 26].

A systematic review and meta-analysis reported a rate of 28.1% of different types of ECG abnormalities as atrial or ventricular arrhythmias [27]. William Regan et al. showed ECG changes in a majority of children (67%) [28]. We found ECG abnormalities in 45% of patients, mainly abnormalities of repolarization, and one of the three cases with a type-1 Brugada pattern presented AF on the first day of hospitalization in intensive care. An Ajmaline test 8 months later confirmed Brugada Syndrome (not yet published). All patients at discharge presented a normal ECG, confirming a fast resolution of electrical anomalies and suggesting reversible damage.

In our pediatric cohort, only one Hispanic patient developed a transient ectasia of the main left coronary, which quickly resolved after therapy. There are conflicting descriptions about the data regarding the involvement of coronary arteries; the incidence of coronary artery dilatation/aneurysms varies significantly among reports (8–24%) [14, 29], and the pathological mechanism is, as yet, unknown. Abrams et al. showed some degree of association between muco-cutaneous alterations, conjunctival hyperemia, and coronary artery alterations, suggesting a possible overlap between MIS-C and Kawasaki disease [22]. None of our 23 patients with muco-cutaneous manifestations had coronary involvement; many factors could have contributed to this difference, such as a strict selection of patients (all showed serological positivity of IgG), a short interval between the onset of symptoms and therapy (median delay of 5 ± 2 days), combined therapy with steroids and intravenous immunoglobulin in almost patients (> 90%), and the prevalence of Caucasian ethnicity. Finally, the accurate recognition of patients with MIS-C may have limited overlap with Kawasaki disease.

We divided the patients in two groups according to severity of cardiac dysfunction on admission, to assess if there was any relationship between clinical and laboratory factors. All our patients had an alteration in the inflammatory index, with CRP, white blood cell, and neutrophil counts significantly higher in patients with more severe cardiovascular involvement; this suggests that myocardial involvement could be the hallmark of a hyperinflammatory state [5].

As expected, cardiac laboratory indices, including troponin T and NTproBNP levels, were higher in group A. Furthermore, values out of the ranges were also found in group B, suggesting that an underestimation of cardiac injury was not excluded using only echocardiographic evaluation.

Interestingly, the median value of NTproBNP was very high in group B as well. It is known that NTproBNP is associated with myocardial damage, stress, and fibrosis, and high NTproBNP levels constitute an independent diagnostic criterion for heart failure. However, it is also known that inflammation appears to be associated with natriuretic peptide release [30, 31]. Increasing NTproBNP in MIS-C without severe cardiac involvement may be related to the significant inflammatory state. Two patients in our series with shock at presentation showed a normal EF (52–60%) and troponin (49–46 ng/L), and had very high NTproBNP values (17.810–23.748 ng/L) and PCR (456–500 mg/L). As reported, shock and hemodynamic compromise in MIS-C can occur with preserved cardiac function.

We noted that patients with more compromised heart function spent more time in resuscitation, however there were no differences between the groups in terms of mortality, ventilation support, and hospital stay. Additionally, LVEF recovery in group A patients did not differ from that in group B, indicating that the cardiac impairment at admission did not affect the therapeutic response. Even though, according to our protocol, more severe patients were treated with high-dose corticosteroids, the limited number of patients did not allow us to estimate the real effect of the different dosages.

Despite significant differences in clinical presentation and the need for intensive care, MIS-C patients with significant cardiac involvement completely recovered from illness without differences in term of cardiac function and length of hospital stay. There is a strict relationship between inflammatory state and cardiac impairment. We confirmed that the heart is one of the main target organs; however, when properly treated and supported in the acute phase, the cardiac involvement does not appear to affect the prognosis.