• 
• 

Three Part Question

In [a patient diagnosed as having Myocarditis] do [steroids improve outcome] in [terms of morbidity and mortality]

Search Strategy

Medline 1966-2014 using MeshPubMed interface
(( "Steroids/administration and dosage"[Mesh] OR "Steroids/adverse effects"[Mesh] OR "Steroids/analysis"[Mesh] OR "Steroids/statistics and numerical data"[Mesh] OR "Steroids/therapeutic use"[Mesh] OR "Steroids/therapy"[Mesh] )) AND ( "Myocarditis/analysis"[Mesh] OR "Myocarditis/drug therapy"[Mesh] OR "Myocarditis/mortality"[Mesh] OR "Myocarditis/statistics and numerical data"[Mesh] OR "Myocarditis/therapy"[Mesh] )

Search Outcome

Search limited to English language papers only. 289 results using Medline with 12 relevant citations. A search of the Cochrane Library identified a further relevant Cochrane review.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Chen et al 2013 China	Eight randomised controlled studies (with 719 participants) were included. These studies compared corticosteroids to no intervention, placebo, supportive therapy, antiviral agents or conventional therapy. Two review authors extracted data independently and results were presented as risk ratios and mean differences, both with 95% confidence intervals.	Meta-analysis of 8 RCTs	Viral detection	Performed on 37% of patients of which 56% were positive	Trials included were small in size and methodology was poor.
			Mortality	Non-significant mortality between corticosteroid groups and control groups (RR 0.93, 95% CI 0.70 to 1.24)
			Left ventricular ejection fraction at 1 to 3 months follow up	Improved in corticosteroid group compared to control (mean difference 7.36%, 95% CI 4.94 to 9.79) but significant variation. New York Heart Association class and left ventricular end-stage systole diameter were unaffected.
			Cardiac enzyme	Creatinine phosphokinase and isoenzme of creatinine phosphate MB were reduced in the corticosteroid group compared to the control group (although there was small participation numbers)
			Adverse events	Insufficient data
Cilliers et al 2012 Africa	Eight randomised control studies involving 996 patients were included. Steroidal agents (including corticotrophin, cortisone, hydrocortisone, dexamethasone and prednisolone) and intravenous immunoglobulin were compared to aspirin, placebo or no treatment.	Meta-analysis of 8 RCTs	Cardiac disease	No significant difference in risk of cardiac disease between corticosteroids and aspirin (relative risk 0.87). Neither prednisolone nor immunoglobulins conferred benefit compared to placebo in more severe valvular disease.	The trials involved were small in size with varied populations and methodology was poor.
			Adverse events	Side effects were only described in three of the trials. These included: weight gain, moon face, buffalo hump, stria and acne with steroids. Nausea, emesis and tinnitus were reported with aspirin.
Aziz et al 2010 Pakistan	Paediatric population (3.7 years +/- 2.9 years) of 178 were clinically diagnosed with acute myocarditis. This was confirmed with chest radiograph, echocardiogram and electrogram. The causes of fever in the tropics (e.g. typhoid, rheumatic fever, malaria, typhoid, diphtheria, collagen diseases and hepatitis) were excluded. The inclusion criteria included acute myocarditis of 3 months duration. 68 children were available for randomisation into a prednisolone group of 44 and a control group of 24. The follow-up period in prednisolone group was 15.1 +/- 9.2 months and in the control group 13.6+/- 10.6 months. The treatment period was for 3 months with prednisolone.	Monocentric, prospective randomised control study	Viral serology	33 children (77.6%) in the prednisolone group and 19 (61.3%) in control group had virology studies. 22 (73.7%) in the prednisolone group and 9 (47.4%) in the control group had positive titres for viral antibodies.	Relatively small numbers of participants and variable follow up lengths between patients.
			Ejection fraction	At one month, the prednisolone group increased ejection fraction from 30.3+/-7.7% to 41.9+/-14.3%, whereas the control group ejection fraction did not increase significantly. Longer follow up data suggested continued improvement in the prednisolone group compared to the control, but participant numbers were low (36 in the prednisolone group and 17 in the control).
Gagliardi et al 2004 Italy	114 consecutive patients with newly diagnosed dilated cardiac myopathy were divided into three groups: A) acute myocarditis (35 children), B) borderline myocarditis (35 children) and C) non-inflammatory cardiomyopathy (44 children). Supraventricular arrhythmias and metabolic disorders were excluded by conventional laboratory tests. All participants had anatomically normal hearts. All patients underwent echocardiography and endomyocardial biopsy. All patients received standard circulatory support, digoxin, furosemide, captopril and warfarin. Groups A and B received cyclosporine and prednisolone for one month then further 6 months with a tapering dose.	Non-randomised prospective study	Mortality	Event free survival was 0.96 at one year and 0.83 after thirteen years in groups A and B. Group C survival at one year was 0.61 and 0.32 after 13 years. Reduced left ventricular ejection fraction, increased left ventricular end diastolic volume and presence of non-inflammatory dilated cardiac myopathy were independent risk factors for earlier adverse events.	No matched control group. Relatively small numbers within groups.
			Left ventricular function	Group A and B showed complete or partial recovery. Worse prognosis in group C.
			Endomysial biopsy complications	One child developed a pericardial effusion, which fully resolved with no further complications.
			Recurrence of congestive heart failure	3 children in group A and 5 in group B had recurrence three months after termination of immunosuppression.
			Side effects	No bacterial or viral infections were identified during the treatment cycle. No children had hypertension or hyperglycaemia. Hirsutism was commonly recorded but receded after termination of treatment.
Frustaci et al 2003 Italy	652 patients between 1997 and 2000 underwent an endomyocardial biopsy. A diagnosis of active myocarditis was made in 112 of which 80 had left ventricular dysfunction, 28 had ventricular arrhythmias and 4 with unexplained chest pain. Among the 80 patients with acute myocarditis and left ventricular dysfunction, 41 (29 men and 12 women with a mean age of 42.9+/13.5 years) were characterised by lymphocytic myocarditis and chronic (>6 months) heart failure unresponsive to conventional therapy. These patients were treated with azathioprine and prednisolone.	Retrospective study	Response to immunosuppression	21 responded to immunosuppression with improvement in NYHA class, reduction in cardiac volumes, and improved function at 6 month and one year follow-up. 20 patients failed to improve: 12 remained stable, 3 underwent cardiac transplantation and 5 died within 6 months after immunosuppression. In responders, a decline in heart rate, loss of gallop rhythm, increased QRS voltages and improved ECG repolarisation anomalies occurred within one week of treatment. In the 3 HCV-positive responders, cardiac function deteriorated a few months after termination of immunosuppression. This was correlated with histological evidence of relapsing myocarditis and improved with recommencing immunosuppression	No control group. Small study group with potential selection bias.
			Side effects	6 patients had increased body weight of >5 kg and 7 developed mild hypertension.
Frustaci et al 2002 Italy	48 patients (32 men and 16 women with a mean age of 45+/-14.9 years) with clinical and histological diagnoses of myocarditis were tested for HCV and other common cartiotropic viruses. 3 patients had anti-HCV antibodies. These individuals were treated with azathioprine and prednisolone.	Monocentric retrospective study	Histology	Lymphocytic myocarditis with myocytes positive for TORDJI-22 and cardiac auto-antibodies were present in all cases. Nested PCR demonstrated negative and positive HCV RNA strands in serum and myocardium.	Very small study group with potential selection bias.
			Response to immunosuppression	Patients treated with prednisolone and azathioprine for six months showed recovery of cardiac volumes and function. At 4 weeks, the control biopsy found that the myocarditis had progressed to a healed phase although HCV RNA was still detectable. Cardiac improved continued at 12-month follow-up.
Herdy et al 1999 Brazil	Of 120 children with rheumatic fever between 1986 and 1998, 70 cases (with 76 episodes) with active carditis treated with methyprednisolone were studied. The modified Jones criteria were used to diagnose rheumatic fever. Prior to corticoid therapy, eradication of bacterial or parasitic infections and tuberculin tests were performed	Monocentric, prospective cohort study	Mortality and morbidity	Three children died during the pulse therapy, and one many years later during mitral valve replacement. Most morbidity was attributable to mitral valve pathology.	Small study group without a control group
Kleinert et al 1997 Australia	29 consecutive children with acute dilated cardiomyopathy underwent endomyocardial biopsy and were divided into three groups based on the histology: Group I (9 children) with “definite” myocarditis, Group II (two children) with “borderline” myocarditis and Group III (18 children). Group I were treated with cyclosporine and prednisolone.	Non-randomised prospective study	Histology	Myocardial inflammation was present in 38% of patients in Group I and II. There was no clinical, electrographic or echographic features to distinguish the patients.	Small study without a control group
			Response to immunosuppression	All Group I patients regained normal left ventricular function compared to 4 of 18 group III patients. Two patients had biopsy-proven relapse after withdrawal of therapy that resolved after recommencing immunosuppression therapy.
Mason 1996 USA	Patients with congestive heart failure of unknown aetiology underwent endomyocardial biopsy and radionuclide measurement of left ventricular ejection fraction. Those with positive histology and LVEF of <0.45 were asked to consent to be enrolled and randomised. 111 patients were recruited and divided into three groups: azathioprine and prednisolone (19 patients), cyclosporine and prednisolone (45 patients), and no immunosuppression (47 patients). After six months, the immunosuppression was stopped and after one year rigid control of conventional therapy was discontinued.	Multi-centric randomised control study	Left ventricular ejection fraction	Both groups LVEF improved significantly by the end point of 28 weeks and continued through week 52. However, there was no difference in LVEF improvement between the two groups or at any other point. More severe disease at presentation correlated with more severe disease at follow-up.	Small study group and complete data analysis did not distinguish between immunosuppressive regimes.
			Mortality	There was no difference in the incidence of all-cause death between the groups. Three variables were multivariate predictors of death: lower LVEF, more intense conventional therapy and a higher CD2+ T-cell count.
Mason et al 1995 USA	Patients with congestive heart failure of unknown aetiology underwent endomyocardial biopsy and radionuclide measurement of left ventricular ejection fraction. Those with positive histology and LVEF of <0.45 were asked to consent to be enrolled and randomised. 111 patients were recruited and divided into three groups: azathioprine and prednisolone (19 patients), cyclosporine and prednisolone (45 patients), and no immunosuppression (47 patients). After six months, the immunosuppression was stopped and after one year rigid control of conventional therapy was discontinued.	Multi-centric randomised control study	Left ventricular ejection fraction	The mean change in LVEF at 28 weeks did not differ significantly between the group of patients who received immunosuppression (gain of 0.10) compared to the control group (gain of 0.07). LVEF at baseline, less intensive conventional drug therapy and shorter duration of disease were positive independent predictors of LVEF at week 28 weeks.	Relatively small study group. Steroids were not investigated as a single therapy.
			Mortality	There were 34 deaths and 10 cardiac transplants among the 111 patients. The immunosuppressino groups or control group did not differ in survival at one year. Base-line LVEF positively correlated with survival and the intensity of conventional therapy negatively correlated with survival.
			Adverse drug effects	Raised creatinine levels by at least 0.5 mg/dL during the therapy were more common in the cyclosporine and prednisolone group (46%) than in the azathioprine and prednisolone group (16%) or the control group (9%). The incidence of new hypertension or severe infection did not differ between the groups.
Camargo et al 1995 Brazil	Among 68 consecutive patients with severe heart failure due to dilated cardiomyopathy, 43 had active myocarditis based on endomyocardial biopsy findings. 19 were boys and 24 were girls, aged 10 months to 15 years (mean of 2.9 years). All patients presented with tachypnoea, tachycardia, sweating, gallop rhythm, pulmonary wheezing, rales and hepatomegaly. Patients were divided into four treatment groups depending on hospital admission sequence: I) controls (received conventional therapy); II) prednisolone (with conventional therapy); III) azathioprine (with conventional therapy and steroids); and IV) cyclosporine (with conventional therapy and prednisolone). Follow-up was for a mean of 8 months.	Partially-randomised control study	Clinical improvement (disappearance of tachypnoea, rales, tachycardia, sweating and hepatomegaly)	2 of 9 in control group; 3 of 12 in prednisolone group; 13 of 16 in azathioprine group; 10 of 13 in cyclosporine group.	Very small groups within study. Patients were moved between groups during the study, as such it is unclear how to include these patients in data interpretation. Incomplete randomisation and potential selection bias.
			Mortality	2 patients in the prednisolone group and one in both the cyclosporine and azathioprine groups.
			Haemodynamic response	There were mild improvements in haemodynamic parameters in the control and prednisolone groups, but significant improvements in both the azathioprine and cyclosporine groups.
			Histology	One in four patients in the control group showed histological improvement. 2 of 5 patients in the prednisolone group showed improved. All 6 patients in the azathioprine group and all 4 in the cyclosporine group had no signs of myocardial inflammation at the end of treatment.
Maisch et al 1994 Germany	38 patients with an infiltrate and haemodynamic impairment (ejection fraction <55% and cardiomegaly) were included. They were randomised into two groups: Group 1 was treated with restraint of physical activity, optional ACE inhibitor, digitalis and diuretics; and Group 2 was treated for 4 weeks with azathioprine and prednisolone then a tapering dose for two months in addition. Follow-up biopsies and haemodynamics were completed after 3 months.	Randomised control study	Biopsy serum	Group I: infiltrate present in 21%, IgG and C3 present in 95%, virology positive in 42%. Group 2: infiltrate present 15%, IgG and C3 in 26% and virology positive in 36%.	A small number of patients were involved in the study with a limited follow-up period. Query: statistical misrepresentation on the results table (NYHA class percentages add up to 110% for Group 1).
			NYHA class	Group I: 21% improved, 21% deteriorated, 68% unchanged. Group II: 53% improved, 24% deteriorated and 26% unchanged.
			Left ventricular ejection fraction	Group I: 15% improved, 26% deteriorated and 59% unchanged. Group II: 47% improved, 26% deteriorated and 32% unchanged.
Haffejee & Moosa 1990 South Africa	35 consecutive children with rheumatic carditis, diagnosed using clinical criteria and excluding patients where the diagnosis was in doubt. Follow-up period was 7 years. The children were divided into two groups: prednisolone group and placebo group.	Randomised control study	Mortality	No significant difference in the short-term or long-term follow up.	Small study group. Only one dosage regime was used for prednisolone.
			Morbidity	There was no significant difference, including the later requirement of valvular surgery. In addition, one third of cases improved with time regardless of treatment. The more favourable outcome was associated with initial mild to moderate disease.

Comment(s)

The methodology within studies was overall poor and the samples sizes were small. There was considerable heterogeneity between studies. It is unclear whether certain regimes could still be of benefit as much of the data is old and the supportive or “conventional” therapies have developed over time. The advent of echocardiography has facilitated better evaluation of cardiac function, but the antiquity of the literature has meant that echography has been underused in the evaluation of cardiac function following myocarditis. The statistical significance of existing literature supporting the use of steroids is weak, while the adverse effects of steroids are well documented. The disappointing results were similar between types of myocarditis (viral, acute with pre-existing cardiomyopathy and rheumatic). Therefore, there is very little evidence to support the use of steroids in the management of myocarditis.

Clinical Bottom Line

There is very little evidence to support the efficacy of steroids in myocarditis.

References

1. Chen et al Corticosteroids for viral myocarditis. Cochrane Database Sytematic Review
2. Cilliers et al Anti-inflammatory treatment for carditis in acute rheumatic fever. Cochrane Database Sytematic Review 2012
3. Aziz et al Acute viral myocarditis: role of immunosuppression: a prospective randomised study. Cardiol Young 2010
4. Gagliardi et al Long term follow up of children with myocarditis treated by immunosuppression and of children with dilated cardiomyopathy. Heart 2004
5. Frustaci et al Immunosuppressive therapy for active lymphocytic myocarditis: virological and immunologic profile of responders versus nonresponders. Circulation 2003
6. Frustaci et al Lone hepatitis C virus myocarditis responsive to immunosuppressive therapy. Chest 2002
7. Herdy et al Rheumatic carditis treated with high doses of pulsetherapy methylprednisolone. Results in 70 children over 12 years. Arg Bras Cardiol 1999
8. Kleinert et al Myocarditis in children with dilated cardiomyopathy: incidence and outcome after dual therapy immunosuppression. J Heart Lung Transplant. 1997
9. Mason Immunopathogenesis and treatment of myocarditis: the United States Myocarditis Treatment Trial. J Card Fail. 1996
10. Mason et al A clinical trial of immunosuppressive therapy for myocarditis. The Myocarditis Treatment Trial Investigators. N Engl J Med. 1995
11. Camargo et al Favorable effects of immunosuppressive therapy in children with dilated cardiomyopathy and active myocarditis. Pediatric Cardiology 1995
12. Maisch Immunosuppressive treatment in autoreactive myocarditis--results from a controlled trial. Postgrad Med J. 1994
13. Haffejee & Moosa A double-blind placebo-controlled trial of prednisone in active rheumatic carditis. Ann Trop Paediatr. 1990