Three Cases of Myocarditis in Childhood Associated with Human Parvovirus (B19 Virus)
- First Online:
- Cite this article as:
- Munro, K., Croxson, M., Thomas, S. et al. Pediatr Cardiol (2003) 24: 473. doi:10.1007/s00246-002-0173-9
- 77 Views
We report three childhood cases of myocarditis associated with human parvovirus (B19 virus). All three children presented with significant cardiac decompensation, with one requiring extracorporeal membrane oxygenation support. Left ventricular function was severely impaired in all three. Myocardial biopsy confirmed histological myocarditis and was positive for B19 virus by nested polymerase chain reaction. Serum was positive for IgG B19 virus but negative for IgM in all three cases. All three children were treated with diuretics, ACE inhibitors, and immunosuppression. Prednisone and cyclosporin were continued until there was echocardiographic and histological improvement. All made a full clinical and echocardiographic recovery.
Human parvovirus B19 (B19 virus) can cause a wide range of human diseases, including erythema infectiosum (fifth disease), arthritis, aplastic anemia, and fetal hydrops. There have been few reports of B19 virus causing myocarditis in nonimmunocompromised patients [2, 5, 6]. We present three cases of myocarditis associated with B19 virus in children.
A 2-year-old boy was admitted with oliguria, vomiting, abdominal pain, and tachypnea. There were no prodomal viral symptoms. Examination revealed basal crepitations, hepatomegaly, a 1/6 systolic murmur, and a gallop rhythm. Chest x-ray showed cardiomegaly and pulmonary congestion. Echocardiogram (ECG) showed a dilated, thin-walled, poorly contracting left ventricle. The left ventricular end diastolic dimension (LVEDD) was 5.3 cm [z score for body surface area (BSA), 8.41], left ventricular end diastolic dimension (LVESD) was 4.3 cm, fractional shortening (FS) was 13%, and ejection fraction (EF) was 25%. The low-voltage ECG with minor ST changes in V1, V5, and V6 supported the clinical diagnosis of inflammatory myocarditis. Our unit’s standard protocol of investigations for dilated cardiomyopathy, includes viral serology [Epstein–Barr virus (EBV), cytomegalovirus (CMV), rubella, mumps, coxsackie, adenovirus, enteroviruses, and B19; serum polymerase chain reaction (PCR) for EBV, CMV, B19 virus; nasopharyngeal aspirate, throat swab, and stools for enteroviruses; and urine for CMV], full blood count, urea and electrolytes, thyroid function tests, CPK, ESR, acid–base balance, ammonia, lactate, carnitine, and urine for amino and organic acids. Hemoglobin was 118 g/L, white blood cell count (WBC) was 8.2, with lymphocytes of 6.4. The heart failure was treated with frusemide, captopril, and digoxin. Cyclosporine and prednisolone were started due to the high clinical suspicion of a lymphocytic myocarditis. A myocardial biopsy 7 days after admission showed excessive lymphocytes but no myocyte necrosis. Serum was IgG positive for B19 virus but IgM negative. Serum and myocardium were DNA positive for B19 virus on nested PCR. The results of all other investigations were normal. Following 6 months of immunosuppressive therapy, his exercise capacity had returned to normal. Echocardiography now showed mild left ventricular dilatation only, LVEDD 3.4 cm (z score = 1.74), LVESD 2.2 cm, FS 36%, and EF 57%. Repeat cardiac biopsy remained PCR positive for B19 virus DNA.
A 7-year-old girl collapsed at home and was given CPR by her grandmother. Two weeks prior to this presentation she had been seen with a viral illness with red cheeks, and 3 days prior to admission there had been vomiting and new viral symptoms. On arrival at the emergency department, she was conscious but in ventricular tachycardia and was treated with lignocaine and flecainide. Chest x-ray showed gross cardiomegaly and the diagnosis of viral myocarditis was postulated. ECG showed an extremely dilated left ventricle with very poor function, LVEDD 5.2 cm (z score = 5.36), LVESD 4.4 cm, FS 15%, EF 41%, moderate mitral regurgitation (MR), dilated right ventricle, and pulmonary arteries. Despite an amiodarone infusion, attempted cardioversion, and artificial ventilation, she required cardiac massage and extracorporeal membrane oxygenation life support for 4 days. She underwent investigation for cardiomyopathy as outlined for case 1. Hemoglobin was 122 g/L and WBC was 23.8. She received immunosuppressive therapy with one dose of IVIG and then prednisolone and cyclosporine. Clinical recovery was rapid and ECG 11 days after admission showed normal ventricular function, LVEDD 3.5 cm, LVESD 1.7 cm, and FS 51%. Myocardial biopsy at that time revealed lymphocytic myocarditis with evidence of active myocyte necrosis. Serum was B19 virus IgM negative but IgG positive and nested PCR positive for B19 DNA. The cardiac biopsy specimen was positive for B19 virus DNA on nested PCR. All other viral serology was negative as for case 1. At 6 months follow-up cardiac biopsy still showed low levels of B19 virus DNA. ECG showed normal ventricular function, with LVEDD 4.2 cm, LVESD 2.8 cm, FS 33%, and EF 59%.
A 19-month-old boy was admitted in cardiogenic shock following sudden onset of tachypnea and cyanosis. There was a history of a viral exanthem 2 months previously. On admission he was poorly perfused and became inotrope dependent. He had a systolic murmur with basal crepitations and hepatomegaly. ECG showed global impairment of dilated left and right ventricles, LVEDD 4.4 cm (z = 4.62), LVESD 3.6 cm, FS 19%, mild tricuspid regurgitation (TR), and moderate MR. He was investigated as per the standard protocol for dilated cardiomyopathy. WBC was 15.9, lymphocyte was 9.7, and Hb was 81 g/L. He received a blood transfusion. Cardiac biopsy 2 days after admission showed heavy lymphocytic infiltrate with myocyte necrosis, confirming the diagnosis of viral myocarditis. He was treated with prednisolone and cyclosporine and required ventilation for 13 days and inotropic support for 10 days. Serum was B19 virus IgM negative but IgG positive and nested PCR positive for B19 virus DNA. Cardiac biopsy was also positive for B19 virus DNA nested PCR. All other viral serology was negative as for case 1. Cardiac biopsy 6 months after presentation still showed low levels of B19 virus DNA on nested PCR. Nine months after presentation ECG showed good ventricular function, LVEDD 3.1 cm (z = 0.08), LVESD 2.1 cm, FS 32%, and EF 52%.
Serological results are summarized in Table 1.
Myocarditis in childhood is usually of viral origin. We believe there is sufficient evidence to indicate that the etiology was probably B19 virus in our three cases. B19 virus DNA was detected by nested PCR in all three myocardial biopsies soon after admission and at 6-month follow-up. This could indicate that the myocardium was the site of parvovirus replication. Note that the serum was positive for IgG antibody to B19 virus but IgM antibody negative. It is expected that B19 virus IgM persists for 2 or 3 months after primary infection , so it could be postulated that myocarditis is a late manifestation of B19 virus infection, occurring after the IgM antibody level has declined. It is notable that in all cases no other viral agent more commonly associated with myocarditis was isolated. In particular, for cases 1 and 2, myocardial tissue was negative for enterovirus RNA by reverse transcriptase PCR (RT-PCR). There was insufficient myocardial tissue for RT-PCR in case 3. This child was negative for enterovirus RNA on RT-PCR of acute serum and was also negative for enterovirus on culture of stool and throat swab. It is of note that parvovirus persisted in the myocardium at 6-month follow-up when it has cleared from the serum. This reduces the possibility that the presence of parvovirus in myocarditis represents contamination of the myocardium with blood. Although we are aware that parvovirus DNA may persist in a variety of tissues without apparent clinical disease [4, 9], the histology was consistent with the diagnosis of viral myocarditis in all three cases with varying degrees of myocyte necrosis.
It is surprising that parvovirus IgM was negative despite positive IgG, especially for Case 2, who had prodomal symptoms. Lack of IgM antibodies in the presence of other antibody markers indicating parvovirus myocarditis has been found by others [1, 6]. This raises the question of the degree of specificity of the IgM B19 in acute parvovirus infection.
The large case control study from Texas, showing positive parvovirus in 2% of myocarditis and cardiac transplant rejection patients but none of the 200 controls, also supports the conjecture that our cases were caused by parvovirus infection .
Seroprevalence studies reveal that only 2% of children younger than 5 years and 21% of 5- to 9-year-olds have IgG antibody for human parvovirus (B19 virus) . Other case studies have also shown a time delay from several weeks to 6 months between presentation with a rash, joint pain, or influenza-like symptoms and onset of the manifestations of myocarditis [2, 7].
Parvovirus infection, which was previously thought to be inconsequential in those immunocompetent and without hemolytic anemia, has now been found to cause severe, even life-threatening, myocarditis [2, 7, 8]. This is a rare complication of a common illness and suggests either that the cardiac involvement is usually subclinical or that the virus is only mildly cardiotrophic. Certainly, there is increasing evidence to suggest that HPV B19 infects more than just erythroid progenitor cells.
The use of immunosuppressive treatment for myocarditis remains controversial in adults and children. In recent years, we have utilized the immunosuppressive regimen described by Kleinert et al.  for biopsy-proven myocarditis in all children presenting with clinical myocarditis. Like our Australian colleagues, we have been impressed by a positive response in the majority of such children presenting with lymphocytic myocarditis, considerably improving the previously variable prognosis for this condition. This protocol for endomyocardial biopsy-positive cases consists of cyclosporine monitored by serum levels and 2 mg/kg prednisolone for 4 weeks, which is then reduced to a 0.1–0.2 mg/kg maintenance dose. Treatment is continued until LV function is normal and there is no evidence of ongoing myocarditis on follow-up endomyocardial biopsy.
In summary, we described three cases of biopsy-proven lymphocytic myocarditis in childhood associated with B19 virus infection. Severe cardiac impairment resolved in all three cases with immunosuppressive treatment.
We thank Barbara Semb for secretarial assistance.