Abstract
Objective. To investigate early cardiac involvement in patients with polymyositis/dermatomyositis (PM/DM), and to evaluate the risk factors for early cardiac impairment.
Methods. The study population included 46 patients with PM/DM who did not have overt cardiovascular manifestations and 21 age- and sex-matched healthy controls. Traditional echocardiography and tissue Doppler imaging (TDI) were used to evaluate cardiac function in both groups. Clinical characteristics were recorded. Multivariate logistics regression analysis was applied to investigate risk factors for early cardiac impairment in patients with PM/DM.
Results. No significant difference was found between patients and controls by traditional echocardiography. However, compared to controls, PM/DM patients had a significantly lower ratio of early diastolic mitral annulus velocity to late diastolic mitral annulus velocity (Em/Am; 1.23 ± 0.52, 1.79 ± 0.37, respectively; t = −4.485, p < 0.001) and a higher ratio of peak early diastolic transmitral flow velocity to Em (E/Em; 8.26 ± 2.57, 6.76 ± 1.17; t = 3.287, p < 0.05) as found by TDI measurements. There was no significant difference between the TDI variables of patients with PM and DM. The multivariate regression analysis showed that female sex (OR 11.044, 95% CI 1.066–114.357, p = 0.044), late onset (OR 1.157, 95% CI 1.047–1.278, p = 0.004), and duration of disease (OR 1.060, 95% CI 1.008–1.115, p = 0.023) were risk factors for abnormal left ventricular filling pressures.
Conclusion. TDI is useful for detecting early cardiac impairment in patients with PM/DM. Left ventricular diastolic dysfunction is an early feature of cardiac involvement. Female sex, late onset, and long course of disease are 3 independent risk factors for predicting left ventricular diastolic dysfunction in patients with PM/DM.
Polymyositis (PM) and dermatomyositis (DM) are autoimmune systemic diseases characterized by chronic muscle weakness and inflammatory cell infiltrations into skeletal muscle. A number of studies show that cardiac involvement is frequent and has been recognized as the main prognostic factor for survival in patients with PM/DM1,2,3. Thus, early detection of cardiac involvement is crucial in patients with PM/DM. Echocardiography, an excellent noninvasive imaging modality, is recognized as a useful tool for assessing heart structure integrity and function. However, subtle changes in myocardial lesions may not be accurately detected by conventional echocardiography because of the presence of preload and afterload. Tissue Doppler imaging (TDI) estimates the displacement velocity of heart tissue during the cardiac cycle, analyzing signals with low frequency and high amplitude produced by myocardial tissue, which allows a quantitative assessment of systolic and diastolic function. TDI has been used to assess cardiac diastolic function in patients with coronary artery disease, hypertension, and type 2 diabetes mellitus with more sensitivity and specificity4,5,6. This technique has, to our knowledge, not previously been systematically evaluated in the PM/DM population. The aim of our study was to investigate heart function in patients with PM/DM without overt clinical cardiovascular manifestations by means of conventional and TDI echocardiography, and to analyze the risk factors for cardiac involvement.
MATERIALS AND METHODS
Study population
A total of 46 hospitalized PM/DM patients (11 PM, 35 DM) who did not have clinical cardiac symptoms were enrolled from the Department of Rheumatology at China-Japan Friendship Hospital between January 2012 and July 2012. All patients fulfilled the criteria of definitive PM or DM proposed by Bohan and Peter7. Patients who presented with accompanying malignancy, overlap syndromes, or inclusion body myositis were excluded. Patients older than age 50 years were excluded. Patients with previously known congenital heart disease, rheumatic heart disease, hypertension, coronary artery disease, diabetes, dysthyroidism, cardiomyopathy, severe renal disease (Cr ≥ 133 μmol/l), or anemia were also excluded. Additionally, patients with atrial fibrillation and more than mild mitral regurgitation were excluded. Ages ranged from 16 to 50 years, with mean 31.1 (± 12.1) years. The average disease duration was 4.78 months (± 13.19; range 0–84 mo). Patients were matched to 21 healthy control subjects on the basis of age and sex. The age of controls ranged from 20 to 50 years, mean 35 years (± 6.7). Pregnant and lactating women were excluded. The study protocol was approved by the Ethics Committee of China-Japan Friendship Hospital, and written informed consent was obtained from each participant. The study was conducted according to the Declaration of Helsinki (2000).
Conventional echocardiography
Transthoracic echocardiography was performed using commercial ultrasound equipment (Philips iE33) with a 1–5 MHz transducer. Left atrial (LA) diameters, left ventricular (LV) diameters, LV end-diastolic diameters (LVEDD), LV end-systolic diameters (LVESD) and interventricular septa (IVS) were determined from 2-D targeted M-mode echocardiography based on the criteria of the American Society of Echocardiography8. Left ventricle ejection fraction (LVEF) were calculated according to the Simpson method9. Mitral inflow velocities were evaluated by pulse-wave Doppler imaging with the sample volume placed at the tip of the mitral leaflets that can be seen on the apical 4-chamber view. Peak early (E) and peak late diastolic transmitral flow velocity (A) were measured to calculate the ratio of peak E to peak A velocities (E/A; Figure 1). The maximal tricuspid regurgitation velocity was measured by continuous wave Doppler echocardiography in the 4-chamber view, and Bernoulli’s modified equation was used to calculate systolic pulmonary artery pressure (sPAP)10. All measurements were taken on 3 consecutive beats, and the mean values were used.
Mitral inflow velocities were evaluated by pulse-wave Doppler techniques. Sm: Peak myocardial systolic velocity. Em: Peak early diastolic myocardial velocity. Am: Peak late diastolic myocardial velocity.
TDI measurement
TDI measures included peak myocardial systolic velocity (Sm), early myocardial diastolic velocity (Em), and late peak myocardial diastolic velocity (Am; Figure 2). TDI data were obtained from 6 sites in the mitral annulus, including the lateral wall (LVLW) and the posterior septum (LVPS) in the apical 4-chamber view; the anterior wall (LVAW) and inferior wall (LVIW) in the apical 2-chamber view; and the anterior septum (LVAS) and posterior wall (LVPW) in the LV long axis view. All TDI measurements were performed in adherence with the guidelines and standards of the American Society of Echocardiography11 by a single observer blinded to patient/control status. All measurements were taken on 3 consecutive beats, and averages were estimated. The Em/Am and E/Em ratio, the index of ventricular diastolic function, and filling pressure were calculated.
TDI in LVPS of mitral annulus. E: Peak early diastolic transmitral flow velocity. A: Peak late diastolic transmitral flow velocity.
Conventional echocardiography and TDI measurement were performed in the first week of hospitalization.
Clinical characteristics
We made detailed records of clinical information, including age at onset, course of disease (recorded monthly), height, weight, blood pressure, heart rate, muscle strength, muscle disease activity, rash (heliotrope rash, Gottron rash, paraungual erythema, and “mechanics hand”), interstitial lung disease (ILD), esophageal lesions, and Raynaud’s sign. Simple muscle strength was rated on a scale from 0 to 5, according to the manual muscle test proposed by Lovett12: 0 = paralyzed; 1 = visible or palpation evidence of muscle contraction; 2 = movement with gravity eliminated; 3 = movement against gravity; 4 = weakness; and 5 = normal. Muscle disease activity was evaluated according to the Myositis Disease Activity Assessment Tool (Version 2, 2005)13 by physician assessment using a visual analog scale (VAS). For each subject, an experienced rheumatologist drew a vertical mark on a 10-cm line to indicate disease activity: a mark at the left end of the line represented no evidence of muscle disease activity; a mark at the midpoint represented moderate disease activity; and a mark at the right represented extreme or maximum disease activity. Examples of patients whose marks were at the rightmost end of the line include bedridden patients with severe muscle weakness resulting in an inability to perform self-care. High-resolution computed tomography of the lungs was performed to investigate radiographic abnormalities consistent with ILD. ILD was defined as the existence of one of the following abnormalities: parenchymal micronodules and nodules, linear opacities, irregularity of the interfaces between peripheral pleura and aerated lung parenchyma, ground-glass opacities, honeycombing, and traction bronchiectases or bronchiolectases14. The main clinical manifestations of esophageal lesions included dysphagia, nasal or gastroesophageal regurgitation, choking, heartburn, aspiration pneumonia, nasal speech, and hoarseness. If one of these symptoms was present, further examinations were carried out to evaluate the esophageal function, including esophageal manometry, barium-swallow examination, or endoscopic examination. Patients were considered to have esophageal complications if one of these auxiliary examinations yielded abnormal results. Laboratory assessment items included total cholesterol, low-density lipoprotein cholesterol, blood glucose, uric acid, creatine kinase, high-sensitivity C-reactive protein (hs-CRP), and erythrocyte sedimentation rate (ESR), which were performed on the second day of hospitalization.
Statistical analysis
Comparisons between patients with PM/DM and controls and among patient subgroups were performed using independent t tests for continuous variables and Pearson chi-square test for categorical variables. Multivariate logistic regression models were used to assess the relationship between clinical variable data and the presence of ventricular diastolic dysfunction. Analysis was performed using PASW statistics 18. Continuous variables are presented as mean ± SD unless otherwise stated. Two-sided values of p < 0.05 were considered statistically significant.
RESULTS
Baseline characteristics
Baseline characteristics of patients and controls are shown in Table 1. There were no significant differences in age, sex, BMI, blood pressure, and heart rate between patients with PM/DM and controls.
Comparison of demographic information between patients with polymyositis/dermatomyositis (PM/DM) and controls.
Conventional echocardiography
Comparisons of the mean values of conventional echo-Doppler variables of patients with PM/DM and controls are shown in Table 2. There were no significant differences in LA, LVEDD, LVESD, IVS, EF, FS, E/A, and sPAP between the 2 groups. Some subjects in both groups showed complicated mild valvular insufficiency without clinical significance. There were 2 patients with pulmonary artery hypertension, one with left atrial enlargement, and 2 patients with septal dyskinesia.
Comparison of conventional echocardiography variables of patients with polymyositis/dermatomyositis (PM/DM) and controls.
TDI measures
The average Sm of mitral annulus in patients with PM/DM was slower than that observed in healthy controls, but the difference was not statistically significant. However, patients with PM/DM had significantly lower Em and Em/Am of mitral annulus and higher E/Em than the controls (Table 3).
Comparison of tissue Doppler imaging between patients with polymyositis/dermatomyositis (PM/DM) and controls.
Comparison of cardiac function between PM and DM
There was no significant difference between the TDI measures of PM and DM patients (Table 4).
Comparison of tissue Doppler imaging of patients with polymyositis (PM) and dermatomyositis (DM).
Relationship between clinical characteristics and LV filling pressure
We used a mitral annulus that had an E/Em ratio < 8 to define normal LV filling pressures. Patients with PM/DM were divided into 2 subgroups according to E/Em values. Comparisons of clinical characteristics between the 2 subgroups showed that there were significant differences in sex, age at onset, course of disease, and proximal muscle strength of lower extremities (Table 5).
Comparison between clinical characteristics of PM/DM patient subgroups, according to E/Em ratio.
In multivariate regression analysis involving sex, age at onset, course of disease, and muscle strength of lower extremities, the risk factors for abnormal LV filling pressures were female sex (OR 11.044, 95% CI 1.066–114.357, p = 0.044), age at onset (OR 1.157, 95% CI 1.047–1.278, p = 0.004), and duration of disease (OR 1.060, 95% CI 1.008–1.115, p = 0.023).
DISCUSSION
To our knowledge, our study represents the first demonstration that LV diastolic dysfunction is an early feature of cardiac involvement in patients with PM/DM. TDI is a suitable tool for evaluation of ventricular function, particularly at the preclinical stage. Female sex, late onset, and long disease course were 3 independent risk factors for increase of LV filling pressure, which predicts LV diastolic dysfunction in PM/DM.
Recently, rheumatologists have paid more attention to cardiac involvement in PM/DM. Heart abnormalities in patients with cardiac symptoms include conduction abnormalities, heart failure, valve disease, pericardial effusions, and pulmonary hypertension, all of which usually have a poor prognosis1. Studies on cardiac complications have not limited the age of patients nor excluded diseases that can affect cardiac function, such as hypertension, diabetes, and dysthyroidism. To detect early heart abnormalities resulting from myopathy and to exclude the possible influence of age, we selected patients under age 50 years without cardiovascular symptoms. Patients who had other accompanying diseases that could influence heart structure and function were also excluded from our study. However, the low sensitivity of conventional echocardiography can limit early detection of subclinical heart abnormalities. Our study shows that there were no significant differences of the conventional echo-Doppler measures between our patients with PM/DM and the control group. A possible reason for the limitations of conventional echocardiography is that conventional echocardiography evaluates systolic function by LVEF and left ventricular fractional shortening in the short-axis view. However, the long-axis dynamics of the human left ventricle are more important in detecting heart dysfunction15. On the other hand, traditional echocardiographic assessment of LV diastolic function relies on Doppler patterns of mitral inflow. Because mitral inflow patterns are highly sensitive to preload and can change dramatically as diastolic dysfunction progresses, the ratio of E/A may show a pseudonormal mitral pattern16. As a result, conventional echocardiography is limited by poor sensitivity and accuracy in the assessment of subclinical ventricular systolic and diastolic dysfunction.
TDI, which was developed in the last decade, estimates the displacement velocity of heart tissue during the cardiac cycle, analyzing signals with low frequency and high amplitude that are produced by the myocardial tissue. It has been shown to be a superior, more sensitive, and more specific method than traditional echocardiography for diagnosis and quantification of systolic and diastolic dysfunction17. Measurement of Sm in the mitral annulus can be used for evaluation of left systolic function. As well, Em and the Em/Am ratio in the mitral annulus can indicate left diastolic function without preload dependence, which can effectively identify a pseudonormal mitral pattern. Simultaneous cardiac catheterization and echocardiographic studies have shown that LV filling pressures correlate with the ratio of the mitral inflow E wave to the tissue Doppler Em wave (E/Em). The E/Em ratio can be used to estimate LV filling pressures as follows11,18: E/Em < 8 correlates with a normal LV end-diastolic pressure. Increased LV end-diastolic pressure indicates impaired LV relaxation, which can aid the detection of early LV diastolic dysfunction.
Using TDI technology, our study showed that there was no significant difference between the average Sm of mitral annuli in patients with PM/DM and in healthy controls. By contrast, patients with PM/DM had significantly lower Em and Em/Am values and higher E/Em values in their mitral annulus than control subjects. These results indicated that patients with PM/DM are more likely to have complications of diastolic dysfunction, which are not easily detected by conventional echocardiography. TDI could become a valuable tool for screening subclinical diastolic dysfunction in patients with PM/DM. Figure 1 and Figure 2 show images of conventional echocardiography (mitral flow E/A > 1) and TDI (posterior septa Em/Am < 1 of the mitral annulus) for the same patient, respectively.
Analysis of the PM and DM subgroups showed that there was no significant difference between their TDI measures. Multivariate regression analysis showed that female sex (OR 11.044, 95% CI 1.066–114.357, p = 0.044), late onset (OR 1.157, 95% CI 1.047–1.278, p = 0.004), and duration of disease (OR 1.060, 95% CI 1.008–1.115, p = 0.023) were 3 independent risk factors that could predict abnormal LV filling pressures in patients with PM/DM. But this early cardiac impairment did not correlate with other clinical characteristics, such as muscle strength, ILD, and blood analyses. In addition, because we excluded patients over age 50 years, the term “late onset” is relative. Our results are not the same as those of previous studies. Sharratt, et al19 and Kehoe, et al20 confirmed an important association between severity of muscle damage and heart involvement. In contrast, Haupt and Hutchins21 and Stern, et al22 both state that there is no relationship between severity of muscle damage and heart abnormality. The reason for this discrepancy might be the cohorts of different patients and/or variations in diagnostic methods. Both possibilities merit further investigation.
Some limitations of our study should be acknowledged. The main limitation of TDI is its dependence on angles6. To minimize error, 3 cycles were measured consecutively, and an average was calculated. Second, PM and DM are diseases with heterogeneity, and the sample size has a substantial effect on the results. An increased number of cases could help clarify cardiac involvement in patients with PM/DM. In addition, there was no significant difference between systolic functions of patients with PM/DM and control subjects. However, the mean duration of disease was relatively short (4.78 ± 13.19 mo). Systolic function could be affected by disease progression and is likely abnormal in a population of patients with PM/DM who have longer disease duration. These possibilities require further investigation as well. Another limitation is that our study was cross-sectional and lacked followup. Our future studies will include followup to clarify the association between disease duration and heart dysfunction.
TDI is currently a safe, noninvasive method to assess heart function in patients with PM/DM, allowing early detection of subclinical LV diastolic dysfunction. Female sex, late onset, and disease duration are predisposing factors for abnormal LV filling pressures, which in turn predict LV diastolic dysfunction. Clinicians should closely monitor the LV diastolic function of patients who have the aforementioned risk factors.
- Accepted for publication May 13, 2013.
