Article Text
Abstract
Objective To study the effects of gender and age on occurrence of myocarditis.
Design Nationwide, multicentre registry study in Finland.
Setting All medical hospital admissions (n=1 698 397) of patients aged ≥16 years during 9.5 years in 29 hospitals.
Patients 3198 myocarditis patients.
Results Myocarditis was more common in men (76.61%; 95% CI 75.11% to 78.05%) than in women (23.39%; 95% CI 21.95% to 24.89%, p<0.0001). Median age of patients was 33 years (IQR 23–50 years). Male patients were significantly younger than females (mean age 34.1±SD 15.1 vs 49.0±18.7 years, p<0.0001). In men, occurrence was highest at 16–20 yrs of age, with a linear decline to elderliness (r=−0.95, p<0.0001). By contrast, myocarditis affected women more evenly at all ages with highest occurrence at the age of 56–60 years. Myocarditis caused 0.19% (95% CI 0.18% to 0.19%) of all medical admissions, and 0.48% (95% CI 0.46% to 0.49%) of admissions due to cardiovascular reasons with an inverse logarithmic association with age (r=−0.97, p<0.0001). Admissions were more commonly caused by myocarditis in men (risk ratio 3.11; 95% CI 2.87 to 3.38, p<0.0001).
Conclusions Men are significantly more susceptible to myocarditis than women. Young men are especially at risk for acquiring myocarditis, while women are affected most commonly at the postmenopausal age. The proportion of hospital admissions caused by myocarditis has an inverse, logarithmic association with age.
Statistics from Altmetric.com
Introduction
Myocarditis is an inflammatory disease of the heart most commonly caused by viral infections.1 Males are found to be at higher risk for viral myocarditis than females in experimental murine studies.2 The majority of patients in reported series have also been men,3 ,4 but these studies have included selected patient populations requiring histological or immunohistological confirmation of myocarditis. Murine studies have also found the susceptibility for myocarditis to be significantly age-dependent with highest sensitivity at adolescence or young adulthood.5 In humans, myocarditis is also commonly regarded as the disease of young adults.1 ,6 However, large-scale studies about the age effect on occurrence of myocarditis are lacking, and current results are conflicting.7 ,8 We studied the effects of gender and age on susceptibility for myocarditis in all-comer adult myocarditis patients using a multihospital, nationwide approach.
Methods
Study patients
All patients ≥16 years of age treated with myocarditis as the primary hospital discharge diagnosis in 29 Finnish hospitals during a period of 9.5 years were included. Patients with cardiac sarcoidosis, cardiac amyloidosis or heart transplant were excluded. The ethnicity of study population was mainly Caucasian.
Data collection
Data of all medical hospital admissions (n=1 698 397) between 1 May 2000 and 31 October 2009 in 29 hospitals were retrospectively collected from the Finnish Hospital Discharge Register (FHDR), a nationwide database containing hospital discharge diagnosis codes (ICD-10) of all hospital admissions in Finland, maintained by the Finnish National Institute for Health and Welfare. Finnish hospital system has three main levels. Five university hospitals represent the highest level in hospital hierarchy followed by 16 central hospitals proving, for example, a wide range of medical and cardiological treatments nationwide. Additionally, smaller regional hospitals treat medical and cardiological patients. This study included all university and central hospitals and eight regional hospitals located across the country. The study was conducted according to the National Institute for Health and Welfare permission THL/1576/5.05.00/2010.
Statistical analysis
Differences between scale variables were analysed with t test, and differences between binomial variables with χ2 test. Correlations between variables were analysed with Spearman’s method. The effects of gender and age on type of myocarditis, aetiology and codiagnoses of patients were analysed using logistic regression models corrected by Firth’s method with both age and gender as main effects. Scale variables are presented as mean±SD, or median with IQR as appropriate. Categorical variables are presented as counts, percentages or risk ratios (RR), with 95% CI as appropriate. Two-sided p values <0.05 were considered statistically significant. The SAS system V.9.3 (SAS Institute, Cary, North Carolina, USA) was used for statistical analyses.
Results
Gender
The study period included 3198 hospital admissions caused by myocarditis. Myocarditis was significantly more common in men (76.61% of patients; 95% CI 75.11% to 78.05%) than in women (23.39%; 95% CI 21.95% to 24.89%, p<0.0001). Median duration of admission was 5 days (IQR 3–6 days). Women had slightly longer duration of hospital treatment compared with men (mean 5.7±5.2 vs 5.2±4.1 days, p=0.027). Type of myocarditis was classified as acute or infectious in 80.52% (95% CI 79.11% to 81.85%) and unspecified in 19.48% (95% CI 18.14% to 20.89%) of patients. Gender did not affect the type of myocarditis.
Age
The median age of all myocarditis patients was 33 years with a range of 16–96 years (IQR 23–50 years) (figure 1A). Male myocarditis patients were significantly younger than females (mean 34.1±15.1 vs 49.0±18.7 years, p<0.0001).The incidence of myocarditis declined linearly after young adulthood in men (r=−0.98, p<0.0001) (figure 1B). By contrast, age distribution of female patients was much more stable, with highest occurrence during the postmenopausal age (years 56–60) (figure 1B). Acute or infectious myocarditis was more common in younger patients (OR 0.93 (95% CI 0.89 to 0.98) per 10 years, p<0.005), while occurrence of unspecified myocarditis increased with age (OR 1.07 (95% CI 1.02 to 1.13) per 10 years, p<0.005).
Aetiology, coinfections and cardiovascular manifestations
Viral aetiology of myocarditis was confirmed during treatment period in 4.44% (95% CI 3.78% to 5.21%) of patients. Myocarditis was caused by bacteria in 2.97% (95% CI 2.44% to 3.62%) and by rheumatic fever in 0.25% (95% CI 0.13% to 0.49%) of patients. The aetiology remained uncharacterised in majority of patients (85.05%; 95% CI 83.78% to 86.25%). Detected aetiology did not differ between genders. Respiratory tract was the most common site of coinfection, followed by otolaryngeal infections (table 1). Age-adjusted distribution of coinfections was similar between genders (table 1). Heart failure was diagnosed in 1.91%, tachyarrhythmia in 1.72% and conduction abnormalities in 0.53% of myocarditis patients (table 1). Of these, women had ventricular arrhythmias and conduction blocks more frequently (table 1).
Myocarditis as the cause of hospital admission
The proportion of admissions caused by myocarditis varied significantly according to age and gender. Proportions of all admissions and cardiovascular admissions caused by myocarditis were inversely correlated with age (r=−0.97, p<0.0001 for both). Myocarditis-caused admissions declined logarithmically with age, but a reincrease in very old (>90 years) patients (figure 2A). Myocarditis caused 0.28% (95% CI 0.27% to 0.29%) of total admissions in men and 0.09% (95% CI 0.08% to 0.10%) in women (RR 3.11; 95% CI 2.87 to 3.38, p<0.0001). Of cardiovascular admissions, myocarditis caused 0.65% (95% CI 0.63% to 0.68%) in male patients and 0.25% (95% CI 0.23% to 0.27%) in female patients (RR 2.57; 95% CI 2.37 to 2.79, p<0.0001). In men, the proportion of admissions caused by myocarditis declined logarithmically with age from young adulthood, while in women the proportion was more a plateau up to 60 years of age (figure 2B).
Discussion
This nationwide, multihospital study describes the effects of gender and age on susceptibility for myocarditis. Although previous studies have reported demographics of myocarditis patients in autopsy series8 ,9 and in patients requiring endomyocardial biopsy,3 ,4 ,10––12 the current report is, to our knowledge, the first large-scale description of gender and age distributions of all-comer adult myocarditis patients.
In our series, 77% of 3198 myocarditis patients were men, and the risk of myocarditis-caused admissions was triple in men compared with women. In a recent report of 222 biopsied myocarditis patients, 72% were men,12 while previous studies have reported somewhat smaller male prominence (60–64%).3 ,4 ,11 Although experimental studies have found potential explanations, the exact mechanisms of higher susceptibility for myocarditis in men are yet unclear.13 Experimental data suggests that although chromosomal differences contribute for gender effect in myocarditis susceptibility, it is dominantly mediated by sex hormones.2 ,13
We found myocarditis to occur in all ages, but young men were especially susceptible. Of all patients, 47% were men aged ≤35, and occurrence of myocarditis was strongly and inversely correlated with age in men. Contrasting with men, the age distribution of female myocarditis patients was notably more stable without increase in young adulthood. In agreement, experimental data shows that testosterone plays a major role in the development of myocarditis. Treatment of female mice with testosterone increases viral replication and inflammation in the heart14 while gonadectomy of male mice significantly reduces cardiac inflammation in experimental coxsackievirus myocarditis.15 Identified mechanisms of testosterone action in myocarditis include increasing viral binding to myocytes,14 commitment to Th1 type immune response,16 inhibition of anti-inflammatory cell populations15 and upregulation of cardiac fibrotic remodelling genes.13 Accordingly, a recent MRI study of acute myocarditis patients found increased fibrosis in men and in patients under 40 years of age,17 and follow-up study of recent onset dilated cardiomyopathy demonstrated significantly better myocardial recovery and prognosis in women.18 Progesterone also increases susceptibility for myocarditis as exogenous progesterone treatment, or infection of female mice in dioestrus phase of menstrual cycle when progesterone levels are naturally elevated, aggravates experimental myocarditis.13 ,19 By contrast, oestrogen plays a protective role in myocarditis by favouring Th2-type immune response,16 stimulating inhibitory T regulator cells and inhibiting proinflammatory T-cells.13 ,20 In line with this, we found the incidence of myocarditis in women to be highest at the age of postmenopausal period when oestrogenic levels are low. Hypothetical mechanism for the gender difference is an effect of exercise. Physical exercise during early phases of viral infection aggravates myocarditis.21 Since women experience more symptoms and report poorer physical health during diseases,22 it is possible that especially young males exercise more commonly than females regardless of signs of viral infection.
Myocarditis caused a significant proportion of cardiovascular and total medical admissions among young adults in the current study. With decreasing incidence of myocarditis and increasing general morbidity with age, the proportion of admissions caused by myocarditis decreased logarithmically, for example, myocarditis was the cause of any medical hospital admission 112 times less likely, and cardiovascular admission 632 times less likely in patients aged ≥80 years of age compared with young adults.
Ventricular arrhythmias are not uncommon in myocarditis patients, and myocarditis is a significant cause of sudden death.6 ,23 Female myocarditis patients had ventricular arrhythmias significantly more frequently in our series. Possible explanation for this is the prolonged ventricular repolarisation which is related to occurrence of ventricular arrhythmias in myocarditis.24 Women are known to have longer baseline QT intervals than men, and prolonged repolarisation is a common feature in myocarditis, present in up to 70% of patients.25 The reduced expression of K-channel subunits and connexin43 in addition to fluctuation of QT interval during menstrual cycle are implicated as causes for prolonged repolarisation and arrhythmia vulnerability in women26 while testosterone is likely to protect men from arrhythmias by shortening action potential.27
This study is limited by the retrospective nature of registry data. However, hospital discharge data has proved to be a valuable information source for documenting the epidemiology of many hospital-treated diseases and completeness and accuracy in the FHDR vary from satisfactory to very good.28 Clinical presentation of myocarditis varies significantly from acute coronary syndrome-type chest pain to arrhythmias and heart failure, making accurate diagnosis a challenging task.6 The golden standard for diagnosis of myocarditis is still endomyocardial biopsy,6 ,23 but due to risk of complications and benefit to only selected patients, biopsy is currently indicated only in limited clinical scenarios.29 MRI is an emerging non-invasive alternative for detection of myocarditis,6 ,23 but restricted availability limits its use in real-life clinical practice. Thus, the diagnosis of myocarditis in clinical practice, and in the current study, is commonly based on studying electrocardiography, cardiac enzymes and echocardiography in addition to exclusion of acute coronary syndromes and clinical judgment. The rate of diagnosed arrhythmias and heart failure in our study was low compared with previous findings of biopsy-confirmed myocarditis patients.4 ,10 This is likely to reflect the different patient populations, but may also be affected by uncertainties in coding codiagnoses.
Since myocarditis is sometimes difficult to differentiate from acute coronary syndrome,30 it is possible that the true occurrence of myocarditis in older patients may be higher than reported here. Other potential pitfalls in diagnosis are recently described stress cardiomyopathy, affecting mainly postmenopausal women, which may be misdiagnosed as myocarditis and differentiation of myocarditis presenting as dilated cardiomyopathy from cardiomyopathies due to other causes.6 In order to maximise the accuracy of real-life diagnosis of myocarditis patients while recruiting as large a patient population as possible, we included only patients who had been examined and diagnosed in hospitals. It is, however, unlikely that this would significantly distort the true gender or age distribution of myocarditis patients.
In conclusion, men are at increased risk of acquiring myocarditis. We found 77% of adult myocarditis patients to be male, and men had a tripled risk for myocarditis-caused admission compared with women. Young men are especially susceptible for myocarditis, while myocarditis affects women most commonly at postmenopausal age. The proportion of hospital admissions caused by myocarditis has an inverse logarithmic association with age.
References
Footnotes
-
Contributors Conception and design, or analysis and interpretation of data: VK, JS and PR. Drafting the article or revising it critically for important intellectual content: VK, JS and PR. Final approval of the version to be published: VK, JS and PR.
-
Funding Study was funded by the clinical research foundation of Turku University Hospital.
-
Competing interests None.
-
Ethics approval Finnish National Institute for Health and Welfare.
-
Provenance and peer review Not commissioned; externally peer reviewed.