Abstract
Acute myocarditis in children is associated with high morbidity and mortality, with limited data on intravenous immunoglobulin (IVIG) treatment and outcome. Our goal was to describe clinical, treatment profile, and predictors of outcome in children with acute fulminant myocarditis (AFM) receiving intensive care. Case records of 120 children with clinical diagnosis of acute myocarditis from January 2008 to December 2018 were analyzed retrospectively. AFM was seen in 89 (74.2%) children of which nearly two-thirds (54 [60.7%]) were hypotensive at admission. The median (interquartile range [IQR]) ejection fraction on echocardiography was 25 (18.5–36%). Eighty-two children (68.3%) received IVIG. Intensive care needs were mechanical ventilation ( n = 71; 59.2%) and inotrope support ( n = 89; 74.2%); median inotrope score being 30 (IQR: 20–55). Twenty-one children died (17.5%). Fever ( p = 0.004), arrhythmia ( p = 0.03), shock ( p = 0.015), higher inotrope score ( p = 0.0001), need for ventilation ( p = 0.025), acidosis ( p = 0.013), AKI ( p = 0.0001), transaminitis ( p = 0.0001), and multiorgan dysfunction ( p = 0.0001) were associated with mortality. The mortality was significantly less in IVIG treated group (12.1 vs. 28.9%; p = 0.02). On multiple logistic regression, MODS ( p = 0.002) was independent predictor of mortality while IVIG treatment ( p = 0.004) was favorably associated with survival. AFM complicated by multiorgan dysfunction carried a poor prognosis. IVIG was associated with survival benefit.
Keywords: acute myocarditis, children, predictors, outcome, intravenous immunoglobulin
Introduction
Acute myocarditis in children is associated with high morbidity and mortality. 1 True incidence of myocarditis in children remains unknown due to lack of standard diagnostic criteria and variable disease manifestations that range from subclinical cardiac dysfunction (asymptomatic) to severe cardiac failure, cardiogenic shock, arrhythmias, or sudden death. 2 A large prospective study, reported approximately 10% of recent-onset, dilated cardiomyopathy was secondary to myocarditis. 3 In a retrospective multicenter study by Soongswang et al, primary myocardial diseases accounted for 1.2% of all cardiovascular diseases of which 27.3% were acute myocarditis. 4 The diagnosis of acute myocarditis is largely clinical; based on acute onset cardiac failure heralded by a viral prodrome and low ejection fraction (EF) on echocardiography (ECHO) supported by elevated biomarkers of myocyte injury (CK-MB [creatine kinase-myoglobin], Troponin T). Histopathological diagnosis is no longer considered gold standard due to caveats which include sampling error (patchy myocardial involvement), interobserver variation, and reduced sensitivity of histopathological findings. 5 Although, high-dose intravenous immunoglobulin (IVIG) and other immunosuppressive agents have been used as therapy since 1991, data on its relation to outcome are limited. 6 Furthermore, studies evaluating their role in children with acute myocarditis have failed to show conclusive results. 7 With this background in mind, we decided to study the clinical profile and predictors of outcome in children with acute myocarditis admitted to our intensive care unit and the role of IVIG in relation to outcome.
Materials and Methods
Case records of children between >1 month and 12 years of age with a clinical diagnosis of acute myocarditis admitted to a Level III pediatric intensive care unit (PICU) of a teaching and referral hospital in North India between January 2008 and December 2018 were analyzed retrospectively after obtaining clearance from the institute ethics committee. Acute myocarditis was diagnosed clinically in children who met the following criteria: (1) acute onset cardiac failure with or without hemodynamic compromise in a child with no preexisting cardiac disease; (2) viral prodrome or flu like illness within 2 weeks before hospitalization; and (3) ECHO suggestive of low EF or fractional shortening with or without dilated cardiac chambers (in whom it could be done). Children with cardiac failure secondary to systemic diseases, sepsis, underlying structural heart disease, chronic dilated cardiomyopathy (DCM), autoimmune diseases (inflammatory bowel disease, Kawasaki's disease, and systemic lupus erythematosus), and patients on cardiotoxic drugs were excluded. Based on their predominant clinical presentation, children were retrospectively categorized into three different cardiac syndromes: (1) acute fulminant myocarditis (AFM): acute heart failure with severe hemodynamic compromise defined as signs of hypoperfusion (weak peripheral and central pulses, delayed capillary filling time >3 seconds, cold peripheries/acrocyanosis, oliguria, metabolic acidosis, and/or decreased consciousness) with mean arterial pressure (MAP) <5th centile for age and sex (hypotensive shock) or maintained MAP >5th centile for that age and sex (compensated shock); (2) acute congestive heart failure (CHF) defined as respiratory distress in absence of lung parenchymal disease, disproportionate tachycardia, muffled heart sounds, and S3 gallop and/or cardiomegaly but without hemodynamic compromise; and (3) dysrhythmia (abnormal rhythm on electrocardiography [ECG] associated with or without AFM or CHF).
Data pertaining to demographic profile, clinical manifestations, progression, hematological (hemoglobin, platelet count), biochemical (ionized calcium, alanine transferase (ALT), aspartate transferase (AST), blood urea nitrogen, creatinine, LDH, C-reactive protein, arterial blood gas parameters, cardiac biomarkers (CK-MB, Troponin T), chest X-ray, ECG, ECHO, treatment (inotrope score, ventilation, IVIG), complications (arrhythmia, multiorgan dysfunction [MODS], cardiac arrest), and outcome (survival or death) were recorded on a predesigned structured proforma. Organ dysfunction was defined according to international pediatric sepsis consensus conference guidelines. 8
Cardio-thoracic ratio and ECG findings (low voltage complexes, ST segment changes, and PR and QT interval duration) were adjusted for age and gender. Left ventricular dysfunction on ECHO was defined as EF <50% and graded as mild (40–49%), moderate (30–39%), and severe (<30%). The IVIG-treated group received the drug within 48 hours of admission at 2 gm/kg over 2 days; majority received the first dose on day of admission itself. Net fluid balance (calculated every 12 hourly as % fluid overload) was maintained between −5 to +10 mL/kg/24 hours either by restricting intake or using furosemide infusion or both as deemed necessary. Outcome was defined as survival or death.
Statistical Analysis
Descriptive statistics were used to analyze demographic profile, investigations, complications, treatment, and outcome (survived or died). Data are presented as mean (SD), median (interquartile range [IQR]), and frequencies as appropriate. For univariate analysis, several variables among survivors and nonsurvivors were compared by using Student's t-test and Mann–Whitney U test for parametric and nonparametric data, respectively, and Chi-square test for categorical data. Variables significant on univariate analysis were subjected to multiple logistic regression to determine the independent predictors of outcome. Odds ratio with 95% CI were calculated for significant variables. A p -value <0.05 was considered significant. Data were analyzed by using IBM SPSS Statistics for Windows, version 22.0.
Results
Baseline Patient Characteristics
The baseline characteristics of the study subjects are as shown in Table 1 . Of the three cardiac syndromes described, 89 (74.2%) had AFM, 24 (20%) had CHF, and 7 (5.8%) had arrhythmia. Twenty-seven (22.5%) among the above had overlapping features of shock and arrhythmia. In the group with AFM, nearly two-thirds (54 [60.7%]) were hypotensive.
Table 1. Baseline characteristics and presenting symptoms ( n = 120) .
| Age (mo), median (IQR) | 4 (2.1–11) |
| Gender ratio (boys:girls) | 1.4:1 |
| PRISM score, median (IQR) | 15 (11.5–20) |
| Presenting symptoms | n (%) |
| Respiratory distress | 112 (93.3) |
| Fever | 83 (69.2) |
| Cough | 81 (67.5) |
| Irritability | 60 (50) |
| Vomiting | 31 (25.8) |
| Palpitations | 4 (3.3) |
| Duration of symptoms, median (IQR) (d) | 3 (2–6) |
Abbreviations: IQR, interquartile range; PRISM, pediatric risk of mortality.
Most common arrhythmia seen was supraventricular tachycardia (SVT; n = 15), followed by ventricular tachycardia ( n = 8), ventricular fibrillation ( n = 2), and one each of multifocal atrial tachycardia and ventricular premature contraction. Seven children had more than one arrythmia. Other ECG abnormalities included ST segment changes ( n = 22), QT and PR interval prolongation ( n = 9) and low voltage QRS complexes ( n = 9). Children with SVT had cardiomegaly (86%) with median (IQR) cardiothoracic ratio of 67 (64–71), shock (86.6%), low EF <50% (81.8%), raised CK-MB (100%), and prolonged requirement of inotropes (mean = 73.14 hours) and ventilation (mean = 96 hours).
Acute kidney injury (AKI) was seen in 33 (27.5%) while 25 (20.8%) had CNS dysfunction (encephalopathy, seizures). More than two organ involvement was seen in 22 (18.3%) children.
Children with AFM had more severe acidosis [median [IQR] = 7.27 [7.15–7.36] vs. 7.4 [7.32–7.41]; p = 0.0001), need for ventilation (69.7 vs. 29%; p = 0.0001), higher proportion of AKI (33.7 vs. 9.7%; p = 0.010), transaminitis (23.6 vs. 0%; p = 0.003), CNS dysfunction (27 vs. 3.2%; p = 0.005), and MODS (24.7 vs. 0%). The median (IQR) EF, however, was similar between those with AFM and without.
The laboratory parameters of our study cohort are as shown in Table 2 . Chest radiograph revealed cardiomegaly in 77 (64.1%) children with median (IQR) cardiothoracic ratio of 65% (60–70). Pulmonary edema was seen in 27.5%.
Table 2. Baseline investigations.
| Laboratory parameters | Median (IQR) |
|---|---|
| Hemoglobin (g/dL) | 9.2 (8.1–10.7) |
| Ionized calcium (mmol/L) | 0.9 (0.57–1.1) |
| pH | 7.30 (7.20–7.40) |
| LDH (U/L) | 1,800 (1,309–3,558) |
| AST (IU/L) Range |
130.5 (62–490.2) (10–12,576) |
| ALT (IU/L) Range |
177 (71.7–509) (23–5,974) |
| BUN (mg/dL) Range |
42 (28–60) (2–198) |
| Creatinine (mg/dL) Range |
0.5 (0.3–0.8) (0.1–3.9) |
| Cardiac biomarkers | |
| CRP (IU/L) Range |
17.4 (6.2–50.9) (0–266) |
| CK-MB (IU/L) Range |
144 (61.5–530) (7–3,143) |
| Troponin T card test a | 15 out of 56 (26.8) a |
Abbreviations: ALT, alanine aminotransferase, AST, aspartate aminotransferase; BUN, blood urea nitrogen; CK-MB, creatine kinase-myoglobin; CRP, C-reactive protein; LDH, lactic dehydrogenase; IQR, interquartile range.
Denotes n (%).
An ECHO could be performed in 105 (87.5%) out of 120 children at the time of admission. Reduced left ventricular EF (LVEF < 50%) was present in majority (91.4%), with median (IQR) EF being 25 (18.5–36%). More than half ( n = 60, 57.1%) had severe ventricular dysfunction. Other echocardiographic findings included ventricular hypokinesia ( n = 74, 70.4%), valvular regurgitation ( n = 46, 43.8%), and dilated heart chambers ( n = 53, 50.5%)
Eighty-nine (74.2%) children required inotrope support; the median (IQR) inotrope score was 30 (20–55). Mechanical ventilation was needed in 71 (59.2%) children. Eighty-two (68.3%) children received IVIG, while 38 (31.7%) did not. Twenty-one (17.5%) children died. The median (IQR) length of PICU stay among survivors was 4.6 (2.8–7.2) days.
Univariate analysis revealed fever, arrhythmia and shock at presentation, higher PRISM score, lower EF, higher inotrope score, need for ventilation, severe metabolic acidosis, transaminitis, AKI, coagulopathy, and MODS were significantly associated with mortality ( Table 3 ). On the other hand, IVIG treatment was associated with significantly better survival (IVIG 87.8% vs. non-IVIG group 71.05%; p = 0.024; Table 4 ). On multiple logistic regression, MODS (odds ratio [OR]: 9.19; 95% confidence interval [CI]: 2.23–37.9; p = 0.002) was independent predictor of mortality while IVIG treatment (OR: 0.094; 95% CI: 0.019–0.47; p = 0.004) was favorably associated with survival ( Tables 3 and 4 ).
Table 3. Comparison of survivors versus nonsurvivors: univariate analysis.
| Predictors | Survivors n = 99 | Nonsurvivors n = 21 | p -Value |
|---|---|---|---|
| Age (mo), median (IQR) | 4 (2–11) | 4 (2.3–7) | 0.803 |
| Gender (boys:girls) | 1.4: 1 | 2:1 | 0.45 |
| PRISM score, median (IQR) | 15 (10.5–19.5) | 18 (14–27.8) | 0.021 b |
| Fever, n (%) | 63 (63.6) | 20 (95.2) | 0.004 a |
| Respiratory failure, n (%) | 69 (69.7) | 16 (76.2) | 0.552 |
| Cardiogenic shock, n (%) | 69 (69.7) | 20 (95.2) | 0.015 a |
| Arrythmia, n (%) | 24 (24.2) | 10 (47.6) | 0.031 a |
| Anemia (Hb g/dl), median (IQR) | 9.4 (8.2–10.7) | 8.7 (7.3–10.6) | 0.185 |
| AST (IU/L), median (IQR) | 108 (60–364) | 844 (139–5,890) | 0.001 b |
| Acidosis (pH < 7.3), median (IQR) | 7.31 (7.2–7.4) | 7.26 (7.07–7.31) | 0.013 b |
| CK-MB, median (IQR) | 135 (58–512) | 202.5 (113.7–541.7) | 0.221 |
| CRP, median (IQR) | 15.8 (5.98–41) | 50 (9.9–111.4) | 0.170 |
| Admission EF, median (IQR) | 25 (20–40) | 23.5 (15–30) | 0.067 |
| Inotrope score, median (IQR) | 25 (20–40) | 60 (41–60) | 0.0001 b |
| Ventilation, n (%) | 54 (54.5) | 17 (81) | 0.025 a |
| IVIG treatment, n (%) | 72 (72.7) | 10 (47.6) | 0.025 a |
| MODS, n (%) | 11 (11.1) | 11 (52.3) | 0.0001 a |
| AKI, n (%) | 18 (18.2) | 15 (71.4) | 0.0001 a |
| CNS, n (%) | 14 (14.1) | 11 (52.4) | 0.0001 a |
Abbreviations: AKI, acute kidney injury; AST, aspartate aminotransferase; CK-MB, creatine kinase-myoglobin; CNS, central nervous system; CRP, C-reactive protein; EF, ejection fraction; IQR, interquartile range; IVIG, intravenous immunoglobulin; MODS, multiorgan dysfunction syndrome; PRISM, pediatric risk of mortality.
p < 0.05 by Chi-square test.
p < 0.05 by Mann–Whitney U test.
Table 4. Comparison of intravenous immunoglobulin versus non-intravenous immunoglobulin treated group.
| Predictors | IVIG ( n = 82) | Non-IVIG ( n = 38) | p -Value |
|---|---|---|---|
| Age (mo), median (IQR) | 4 (2–9.1) | 5 (2.9–30) | 0.113 |
| Gender (boys:girls) | 1.4:1 | 1.5:1 | 0.837 |
| PRISM, median (IQR) | 13 (10.5–19) | 17 (13.3–20.8) | 0.060 |
| Fever, n (%) | 54 (65.9) | 29 (76.9) | 0.248 |
| Respiratory failure, n (%) | 62 (75.6) | 23 (60.5) | 0.091 |
| Shock, n (%) | 61 (74.4) | 28 (73.7) | 0.934 |
| Arrythmia, n (%) | 22 (26.8) | 12 (31.6) | 0.591 |
| Anemia, Hb (g/dL), median, (IQR) | 9.19 (8.14–10.7) | 9.4 (8–10.9) | 0.892 |
| AST (IU/L), median (IQR) | 133.5 (62.7–543.7) | 126.5 (61.2–443.7) | 0.773 |
| CK-MB, median (IQR) | 144 (67–573) | 176 (53.6–440.5) | 0.687 |
| pH, median (IQR) | 7.3 (7.17–7.37) | 7.33 (7.26–7.41) | 0.061 |
| Cardiomegaly, n (%) | 57 (69.5) | 20 (52.6) | 0.073 |
| Admission EF, median (IQR) | 25 (15–31) | 30 (25–40) | 0.004 b |
| Ventilation, n (%) | 47 (57.3) | 24 (63.2) | 0.545 |
| Duration of ventilation (h), median (IQR) | 108 (60–174) | 62 (30–96) | 0.007 b |
| Inotropes, n (%) | 61 (74.4) | 28 (73.7) | 0.934 |
| Duration of inotropes (h), median (IQR) | 72 (48–120) | 60 (36–97.5) | 0.088 |
| IS score, median (IQR) | 30 (10–50) | 42.5 (20–60) | 0.318 |
| MODS, n (%) | 14 (7.3) | 4 (10.5) | 0.600 |
| AKI, n (%) | 20 (24.4)1 | 3 (34.2) | 0.262 |
| Duration in PICU (h), median (IQR) | 110 (64.5–179) | 96 (48–120) | 0.105 |
| Survived, n (%) | 72 (87.8) | 27 (71.05) | 0.024 a |
Abbreviations: AST, aspartate aminotransferase; CK-MB, creatine kinase-myoglobin; CRP, C-reactive protein; EF, ejection fraction; IQR, interquartile range; IVIG, intravenous immunoglobulin; MODS, multiorgan dysfunction syndrome; PICU, pediatric intensive care unit; PRISM, pediatric risk of mortality.
p < 0.05 by Chi-square test.
p < 0.05 by Mann–Whitney U test.
Among the 99 survivors, only 32 (32.3%) came for follow up visits; eight patients (25%) persisted to have ventricular dysfunction and had progressed to DCM.
Discussion
Although acute myocarditis in children is associated with significant morbidity and mortality, data from a developing country are limited. Our study on pediatric myocarditis from a tertiary referral hospital attempts to address this gap. We have described the heterogeneous clinical presentation and predictors of survival in acute myocarditis.
Most of the children in our study were infants unlike the bimodal age distribution reported in some previous studies. 7 This is partly related to our department policy of admitting children ≤12 years under pediatrics. Early diagnosis of acute myocarditis is important as the disease has a tendency to progress rapidly to a life-threatening state, but early suspicion in infants is difficult in absence of cardiac specific symptoms like palpitation and chest pain. Most symptoms in our study were extremely nonspecific, heterogeneous, and mimicked a viral respiratory infection. Belying these nonspecific symptoms, nearly three-fourths of these children presented in shock. Our findings imply that a high index of suspicion for acute myocarditis must be entertained when infants with nonspecific respiratory symptoms present with sudden oncoming shock or heart failure. Presence of arrythmia adds further specificity to the diagnosis of acute myocarditis.
Chong et al in their case–control study reported five potentially discriminating factors for clinical diagnosis of acute myocarditis: respiratory distress (OR: 21.3 95% CI: 2.63–172.4); poor perfusion (OR: 11; 95% CI: 3.67–32.89); hypotension (OR: 12.6; 95% CI: 3.32–48.08); any ECG abnormality (OR: 43.8; 95% CI: 2.49–770.3); and cardiomegaly, pulmonary congestion, or pleural effusion (OR: 5.5; 95% CI: 1.93–15.3) with a positive likelihood ratio of 13 (95% CI: 3.31–51.06) for diagnosis of myocarditis if >3 factors were present. 9 In our cohort too, all five factors were present making the clinical diagnosis of myocarditis highly likely.
Our cohort was far sicker than that reported previously. 1 The presentation was more fulminant, with cardiogenic shock and severe LV dysfunction. This was evident from the much lower median (IQR) EF of our patients (25 [18.5–36%]) as compared with 32 (27–36) of other published studies, 1 higher need for ventilation (59.2%) and inotropes (74.2%) compared with 37.5 and 44.9% reported need for ventilation and inotropes, respectively, by Klugman et al. 10
The role of biomarkers in diagnosis of myocarditis is limited. CRP being an acute phase reactant is neither sensitive nor specific for myocarditis. Elevated semiquantitative CRP levels were seen in a higher proportion of children in our study compared with others. 1 Similarly, CK-MB was elevated in majority of the subjects tested. As opposed to this, elevated troponin T levels were seen in only about one-fourth of those tested. In a multicenter study on pediatric myocarditis, higher levels of troponin I were reported in patients with mild ventricular dysfunction than those with moderate or severe ventricular dysfunction. 7 Although elevated troponin levels can assist in diagnosis, a normal result does not exclude myocarditis. 1 Furthermore, an elevated troponin should not be used for grading severity or prognosis. Elevated transaminases were seen in almost half of our patients, possibly related to shock and CHF or secondary to the same viral trigger that caused myocarditis. One pediatric study found raised serum aspartate aminotransferase in children with myocarditis. 11 ECG and chest X-ray have a limited role in diagnosis. Although an abnormal ECG was seen in nearly three-fourths of our patients, the changes observed were nonspecific and variable, similar to that reported by other studies. 12 13 14 15
The most common arrhythmia seen in our study was SVT. This is in contrast to most other studies where ventricular arrythmias were more common. 16 17 SVT usually presents as an acute emergency, but if unrecognized for a long time, the persistent tachyarrhythmia can lead to cardiac dysfunction (tachycardiomyopathy) and heart failure. In such cases, to determine what came first, the tachyarrythmia or cardiac dysfunction may be difficult . It has also been shown that many children with tachyarrythmia tend to have antecedent history of infection and nonspecific symptoms such as dyspnea, palpitation, and cough which mimic diagnosis of myocarditis and DCM. 18 The improvement in cardiac function over time with normalization in heart rate is diagnostic of tachycardiomyopathy. The median time recovery of cardiac function in tachycardia induced cardiomyopathy is <2 months 19
Children with SVT in our series had features of myocardial dysfunction in the form of cardiomegaly, shock, reduced EF, and raised CK-MB. However, whether SVT was consequent to the myocarditis or vice versa cannot be conclusively stated in the absence of a long-term follow-up and detailed electrophysiological studies to rule out implicit WPW syndrome.
The most useful bedside diagnostic tool in cases with a strong clinical suspicion of myocarditis is ECHO. 7 In our cohort, all children who underwent an ECHO showed echocardiographic alterations; majority had severe LV dysfunction and reduced global LVEF. This finding is in contrast to other studies where most patients had mild–moderate LV dysfunction. 1 Our findings suggest that no single laboratory marker is diagnostic. Combination of tests which include raised CRP, troponin T, CK-MB along with abnormal ECHO findings helps arrive at a clinical diagnosis. ECHO additionally helps in stratifying the severity of myocarditis and ruling out other anatomical/structural causes of heart failure. 20
We believe that differentiating acute myocarditis from acute presentation of DCM on clinical grounds is difficult. Long-term follow-up and genetic testing are required to make a clear cut distinction. In the absence of these, a combination of clinical, biomarker, and ECHO-based findings are used to reach a reasonable conclusion of “probable myocarditis.” 21 22 23
Furthermore in a study by Suthar et al, history of fever, arrhythmia, increased Trop I and CK-MB, absence of left ventricular dilation, segmental wall motion abnormalities, and less severe depression of left ventricular systolic function favored the diagnosis of myocarditis over DCM. 24 In another study by Soongswang et al, CK-MB levels were significantly higher in myocarditis as compared with DCM and controls. 25 In our study, fever was a significant symptom and also a poor prognostic factor on univariate analysis. The degree of elevation of CK MB in our cohort was approximately six times higher than normal limits; both findings highly suggestive of a “probable myocarditis' than DCM.”
Treatment of myocarditis is predominantly supportive. IVIG has both antiviral and anti-inflammatory effects on myocarditis, but its efficacy in improving short-term or long-term outcomes is devoid of a consensus. The use of IVIG in suspected myocarditis therefore varies between pediatric centers. 26 27 A favorable response in ventricular function seen by Drucker et al 26 in patients treated with IVIG did not translate into clear cut survival advantage when compared with historical cohorts (those who did not receive IVIG). 7 Klugman et al in their multicenter trial noted that although a trend toward more frequent use of IVIG in patients with extreme severity scores ( n = 27; 27.6%; p = 0.06), it did not confer a survival advantage to the sickest of children. 10 Of the seven patients with a major severity score who died, two received IVIG (6.9%), compared with five who did not (12.8%; OR: 0.5; p = 0.4). Similarly, 4 out of 27 (14.8%) with extreme severity scores who died received IVIG, compared with 6 out of 20 patients (30%) who did not (OR: 0.4, p = 0.2) 10
As opposed to the above, we observed a survival advantage with IVIG. A recent meta-analysis on the role of IVIG for acute myocarditis in children and adults showed IVIG resulted in better recovery of left ventricular function and lower in hospital mortality. 28 Our findings of survival benefit in IVIG treated acute myocarditis holds promise for resource constrained set ups where absence of ECMO, mechanical cardiac support, and cardiac transplant leaves very limited therapeutic options. However, larger multicenter randomized controlled trials are warranted to substantiate the benefits of IVIG in acute myocarditis.
Twenty-one children (17.5%) in our cohort died; a mortality higher than that reported previously (5–7.8%) expectedly because our cohort comprised of ICU-treated patients with severe left ventricular dysfunction. Our findings compare favorably with that of Klugman et al, wherein 115 out of 216, (53.2%) patients with myocarditis stratified as major or extreme severity and admitted to ICU ( n = 116; 53.7%) had a mortality of 14.7%. 10 Although the overall mortality in their study was 7.8% (17 out of 216), the mortality among those requiring ICU was double (17 out of 116) with severity of illness being the significant predictor of mortality (OR: 7.84; 95% CI: 2.36–26.04). Contrary to this, the study by Rodriguez-Gonzalez et al showed a lower mortality of 5% at hospital discharge possibly because AFM formed a small proportion (12 out of 42 patients) and a better median (IQR) EF of 32 (27–36) suggesting a lesser severity of myocardial dysfunction. 1
Our findings revealed that fever, shock, arrhythmia, AKI, transaminitis, and MODS along with need for ventilation and high inotrope support were associated with mortality. On multivariate analysis, MODS secondary to shock was an independent predictor of mortality. This reiterates the importance of timely and aggressive hemodynamic stabilization in children with fulminant myocarditis to prevent organ dysfunction. It is said that those with fulminant myocarditis are more likely to recover if cardiac support is adequate 20
Only one-fourth of our patients returned for follow-up; reasons for which could not be clearly ascertained. Long-term follow-up studies have reported the development of DCM in 16% of patients. 1 One-fourth of our children who returned for follow-up had persistent ventricular dysfunction and had progressed to DCM. Unfortunately, the long-term outcome could not be studied as a whole in our patients.
Our single center study with a large sample size of AFM treated in an intensive care set up is the first of its kind from a resource limited set up. The study suffers from the inherent limitations of a retrospective analysis. Follow-up data were available only in a small proportion of children, as a result of which analysis of risk factors for long-term outcomes could not be studied.
Conclusion
AFM complicated by MODS carried a poor prognosis. IVIG was associated with survival benefit. IVIG use in similar resource limited set ups, with no provision of mechanical cardiac support or ECMO is a viable option.
Funding Statement
Funding None.
Footnotes
Conflict of Interest None declared.
References
- 1.Rodriguez-Gonzalez M, Sanchez-Codez M I, Lubian-Gutierrez M, Castellano-Martinez A. Clinical presentation and early predictors for poor outcomes in pediatric myocarditis: a retrospective study. World J Clin Cases. 2019;7(05):548–561. doi: 10.12998/wjcc.v7.i5.548. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Feldman A M, McNamara D. Myocarditis. N Engl J Med. 2000;343(19):1388–1398. doi: 10.1056/NEJM200011093431908. [DOI] [PubMed] [Google Scholar]
- 3.Felker G M, Hu W, Hare J M, Hruban R H, Baughman K L, Kasper E K. The spectrum of dilated cardiomyopathy. The Johns Hopkins experience with 1,278 patients. Medicine (Baltimore) 1999;78(04):270–283. doi: 10.1097/00005792-199907000-00005. [DOI] [PubMed] [Google Scholar]
- 4.Soongswang J, Sangtawesin C, Sittiwangkul R. Myocardial diseases in Thai children. J Med Assoc Thai. 2002;85 02:S648–S657. [PubMed] [Google Scholar]
- 5.Baughman K L. Diagnosis of myocarditis: death of Dallas criteria. Circulation. 2006;113(04):593–595. doi: 10.1161/CIRCULATIONAHA.105.589663. [DOI] [PubMed] [Google Scholar]
- 6.Ghelani S J, Spaeder M C, Pastor W, Spurney C F, Klugman D. Demographics, trends, and outcomes in pediatric acute myocarditis in the United States, 2006 to 2011. Circ Cardiovasc Qual Outcomes. 2012;5(05):622–627. doi: 10.1161/CIRCOUTCOMES.112.965749. [DOI] [PubMed] [Google Scholar]
- 7.Butts R J, Boyle G J, Deshpande S R. Characteristics of clinically diagnosed pediatric myocarditis in a contemporary multi-center cohort. Pediatr Cardiol. 2017;38(06):1175–1182. doi: 10.1007/s00246-017-1638-1. [DOI] [PubMed] [Google Scholar]
- 8.International Consensus Conference on Pediatric Sepsis . Goldstein B, Giroir B, Randolph A. International pediatric sepsis consensus conference: definitions for sepsis and organ dysfunction in pediatrics. Pediatr Crit Care Med. 2005;6(01):2–8. doi: 10.1097/01.PCC.0000149131.72248.E6. [DOI] [PubMed] [Google Scholar]
- 9.Chong S L, Bautista D, Ang A S. Diagnosing paediatric myocarditis: what really matters. Emerg Med J. 2015;32(02):138–143. doi: 10.1136/emermed-2013-202926. [DOI] [PubMed] [Google Scholar]
- 10.Klugman D, Berger J T, Sable C A, He J, Khandelwal S G, Slonim A D. Pediatric patients hospitalized with myocarditis: a multi-institutional analysis. Pediatr Cardiol. 2010;31(02):222–228. doi: 10.1007/s00246-009-9589-9. [DOI] [PubMed] [Google Scholar]
- 11.Freedman S B, Haladyn J K, Floh A, Kirsh J A, Taylor G, Thull-Freedman J. Pediatric myocarditis: emergency department clinical findings and diagnostic evaluation. Pediatrics. 2007;120(06):1278–1285. doi: 10.1542/peds.2007-1073. [DOI] [PubMed] [Google Scholar]
- 12.Checchia P A, Kulik T J. Acute viral myocarditis: diagnosis. Pediatr Crit Care Med. 2006;7:S8–S11. [Google Scholar]
- 13.Kern J, Modi R, Atalay M K, Kochilas L K. Clinical myocarditis masquerading as acute coronary syndrome. J Pediatr. 2009;154(04):612–615. doi: 10.1016/j.jpeds.2008.10.018. [DOI] [PubMed] [Google Scholar]
- 14.Batra A S, Epstein D, Silka M J. The clinical course of acquired complete heart block in children with acute myocarditis. Pediatr Cardiol. 2003;24(05):495–497. doi: 10.1007/s00246-002-0402-2. [DOI] [PubMed] [Google Scholar]
- 15.Ichikawa R, Sumitomo N, Komori A. The follow-up evaluation of electrocardiogram and arrhythmias in children with fulminant myocarditis. Circ J. 2011;75(04):932–938. doi: 10.1253/circj.cj-10-0918. [DOI] [PubMed] [Google Scholar]
- 16.Anderson B R, Silver E S, Richmond M E, Liberman L. Usefulness of arrhythmias as predictors of death and resource utilization in children with myocarditis. Am J Cardiol. 2014;114:1400–1405. doi: 10.1016/j.amjcard.2014.07.074. [DOI] [PubMed] [Google Scholar]
- 17.Liberman L, Anderson B, Silver E S, Singh R, Richmond M E.Incidence and characteristics of arrhythmias in pediatric patients with myocarditis: a multicenter study. Morgan Stanley Children's Hospital of New York -Presbyterian, Columbia University Medical Center, New York, USAJ Am Coll Cardiol 2014;63(12):
- 18.Arslan A, Erdoğan İ, Varan B, Yılmaz M, Özin M B, Tokel N K. Reversible cardiomyopathy-tachycardiomyopathy in children. Turk J Pediatr. 2019;61(04):552–559. doi: 10.24953/turkjped.2019.04.012. [DOI] [PubMed] [Google Scholar]
- 19.Moore J P, Patel P A, Shannon K M. Predictors of myocardial recovery in pediatric tachycardia-induced cardiomyopathy. Heart Rhythm. 2014;11(07):1163–1169. doi: 10.1016/j.hrthm.2014.04.023. [DOI] [PubMed] [Google Scholar]
- 20.Tunuguntla H, Jeewa A, Denfield S W. Acute myocarditis and pericarditis in children. Pediatr Rev. 2019;40(01):14–25. doi: 10.1542/pir.2018-0044. [DOI] [PubMed] [Google Scholar]
- 21.Chang Y J, Hsiao H J, Hsia S H. Analysis of clinical parameters and echocardiography as predictors of fatal pediatric myocarditis. PLoS One. 2019;14(03):e0214087. doi: 10.1371/journal.pone.0214087. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Freedman S B, Haladyn J K, Floh A, Kirsh J A, Taylor G, Thull-Freedman J. Pediatric myocarditis: emergency department clinical findings and diagnostic evaluation. Pediatrics. 2007;120(06):1278–1285. doi: 10.1542/peds.2007-1073. [DOI] [PubMed] [Google Scholar]
- 23.Hsiao H J, Hsia S H, Wu C T. Clinical presentation of pediatric myocarditis in Taiwan. Pediatr Neonatol. 2011;52(03):135–139. doi: 10.1016/j.pedneo.2011.03.005. [DOI] [PubMed] [Google Scholar]
- 24.Suthar D, Dodd D A, Godown J. Identifying non-invasive tools to distinguish acute myocarditis from dilated cardiomyopathy in children. Pediatr Cardiol. 2018;39(06):1134–1138. doi: 10.1007/s00246-018-1867-y. [DOI] [PubMed] [Google Scholar]
- 25.Soongswang J, Durongpisitkul K, Nana A. Cardiac troponin T: a marker in the diagnosis of acute myocarditis in children. Pediatr Cardiol. 2005;26(01):45–49. doi: 10.1007/s00246-004-0677-6. [DOI] [PubMed] [Google Scholar]
- 26.Drucker N A, Colan S D, Lewis A B. Gamma-globulin treatment of acute myocarditis in the pediatric population. Circulation. 1994;89(01):252–257. doi: 10.1161/01.cir.89.1.252. [DOI] [PubMed] [Google Scholar]
- 27.Gagliardi M G, Bevilacqua M, Bassano C. Long term follow up of children with myocarditis treated by immunosuppression and of children with dilated cardiomyopathy. Heart. 2004;90(10):1167–1171. doi: 10.1136/hrt.2003.026641. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Huang X, Sun Y, Su G, Li Y, Shuai X. Intravenous immunoglobulin therapy for acute myocarditis in children and adults. Int Heart J. 2019;60(02):359–365. doi: 10.1536/ihj.18-299. [DOI] [PubMed] [Google Scholar]