Electrophysiological, structural, and functional disorders in patients with inflammatory cardiomyopathy secondary to inflammatory myopathy

Abstract Background Inflammatory cardiomyopathy (IC) is a syndrome with chronic myocarditis and cardiac remodeling. This study aimed to explore predicting factors of adverse outcomes in patients with IC secondary to idiopathic inflammatory myopathy (IIM‐IC). Methods By means of a single‐center retrospective study, 52 patients with IIM‐IC at Peking Union Medical College Hospital were identified from January 1999 to June 2019. Electrocardiogram and echocardiography data were analyzed for the primary outcome (defined as all‐cause death) and secondary outcomes (defined as re‐hospitalization of heart failure and all‐cause death), using regression and survival analysis. Results The prevalence of atrial fibrillation, ventricular tachycardia, Q‐wave abnormality, left ventricular conduction abnormalities, and reduced left ventricular ejection fraction (LVEF) (≤40%) were 65.4%, 67.3%, 67.3%, 61.6%, and 50.5%. After a median follow‐up of 2 years (IQR 0.8–3.0), 26 cases were readmitted due to heart failure. Twenty‐two deaths were recorded, including 20 cardiogenic deaths. Among the patients with adverse events, the incidence of poor R‐wave progression, low‐voltage of the limb leads, Q‐wave abnormality, QRS duration >130 ms, left ventricular enlargement, and impaired systolic function were higher. Kaplan–Meier analysis showed that Q‐wave abnormality, limb leads low‐voltage, LVEF ≤40%, and left ventricular end‐diastolic dimension >60 mm were correlated with shorter survival. However, multivariate Cox regression analysis revealed that only Q‐wave abnormality (HR = 12.315) and LVEF ≤40% (HR = 5.616) were independent risk factors for all‐cause death. Conclusion Q‐wave abnormality and reduced LVEF are predictive of poor prognosis in patients with IIM‐IC.


| INTRODUC TI ON
Inflammatory cardiomyopathy (IC) is a clinical syndrome defined as myocarditis concomitant with systolic and/or diastolic dysfunction and ventricular remodeling (Caforio et al. 2013). Myocarditis may be diagnosed clinically, by imaging and abnormal cardiac biomarkers, or pathologically, by endomyocardial biopsy showing inflammatory infiltration and nonischemic myocyte injury (Aretz Ht et al, 1987). IC can be due to a variety of etiologies, including infectious, autoimmune, or idiopathic (Caforio et al. 2013). Autoimmune myocarditis is reported increasingly in patients with rheumatic diseases such as systemic lupus erythematosus, systemic sclerosis, and idiopathic inflammatory myopathies (IIM). Among them, IIM is a series of rare immune-mediated inflammatory diseases mainly including dermatomyositis and polymyositis (Hoogendijk et al., 2004). In IIM, skeletal muscular injury is typical while myocarditis can also be concomitant.
The frequency of electrocardiogram changes could be observed in 72% of IIM patients (Lundberg 2006), and autopsy show that about one-third of cases with IIM have subclinical myocardial changes (Denbow et al. 1979). Moreover, cardiac injury was cited as the primary cause of death in 10-20% of IIM patients (Gupta et al. 2011).
When heart failure (HF) or fatal arrhythmia complications happened, the mortality could reach 46.3% in IIM (Zhang et al. 2012), and the cardiac dysfunction was a predictive factor of poor prognosis in IIM (Marie 2012).
It is valuable to identify electrophysiological abnormalities and subclinical heart impairment in IIM (Mahrholdt et al. 2004) and to recognize patients with IC from patients with primary DCM.
Previous studies reported that the most common cardiac features of polymyositis and dermatomyositis were arrhythmia and diastolic HF (Guerra et al. 2017), while cardiomyopathy and systolic HF were rare (Gupta et al. 2011). In fact, reports of prognostic relevance of electrocardiogram in patients with IIM-associated IC (IIM-IC) are scant.
Here, we presented a retrospective cohort study of IIM-IC patients, analyzed their electrocardiographic and echocardiographic characteristics, and followed up with them to explore the possible predicting factors of adverse outcomes.

| Patient selection
Fifty-two consecutive patients diagnosed with IC and IIM at our hospital between 1999 and 2019 were retrospectively studied.
All the cases were transferred from other hospitals, because of the complexity and difficulty of their diseases. The diagnosis of IC was based on the presence of myocarditis and DCM. Myocarditis was confirmed by typical clinical presentations, newly abnormal electrocardiographic features, elevated cardiac troponins, functional, and structural abnormalities on cardiac imaging, according to the 2013 ESC position statement of myocarditis (Caforio et al. 2013). Definition of DCM was in accordance with the 2007 ESC classification of cardiomyopathies, characterized by left ventricular dilation, and impaired systolic function (Elliott et al. 2008).
Serum level of N-terminal brain natriuretic peptide precursor (NT-proBNP) was higher than 400 ng/L in all involved patients (Yancy et al. 2013). According to medical history, echocardiography, and coronary angiography, patients with familial histories of cardiomyopathy, or other organic heart diseases such as coronary heart disease, congenital disease, or rheumatic heart disease were excluded. Patients who were lost to follow-up and who were concomitant with malignant tumors or chronic kidney disease (defined as estimated glomerular filtration rate < 45 ml/min) were also excluded.
The diagnosis of polymyositis and dermatomyositis was based on the 2017 European League Against Rheumatism/American College of Rheumatology classification criteria for IIM, with a definite diagnosis of IIM according to muscle biopsies, when the total aggregate score was 8.7 or higher corresponding to a probability of at least 90% (Bazzani et al., 2010). Muscle biopsies were assessed in the neuropathological laboratory of PUMCH with accepted histopathological diagnostic criteria (Bazzani et al., 2010). The diagnosis of IIM-IC was confirmed through multidisciplinary consultations of cardiologists, rheumatologists, radiologists, and pathologists. Figure   S1 represents a typical example of electrocardiogram, echocardiography, cardiovascular magnetic resonance (CMR), and result of quadriceps muscular biopsy. The study flow chart is shown in Figure 1.
This study is a part of the study "Diagnostic and imaging indicators of immunocardiomyopathic patients," which has been registered in Clinical trials (NCT 03885375), and is conducted with the approval of the Institutional Review Board of Peking Union Medical College Hospital. Due to the retrospective nature of the study, informed consent is waived.

| Data collection
The baseline clinical data, including demographic data (such as disease onset age and gender), discharge diagnoses, disease duration, clinical manifestations, NYHA cardiac functional classification, and treatment strategies, were recorded. Initial electrocardiographic images and echocardiographic reports during their first hospitalization were analyzed by 2 cardiologists who were blinded of the patient's illness.

| Electrocardiographic and echocardiographic data
The baseline electrocardiographic variables were analyzed: resting heart rate, electronic axis and voltage, QRS duration (normal

| Follow-up and endpoints
Structured surveys were conducted through medical record reviews and outpatient visits, and data were supplemented by family report via phone to capture deaths outside the hospital. All cases were followed up until June 2020. Primary endpoint was defined as all-cause death, and secondary endpoints were defined as adverse events including all-cause death and HF requiring re-admission. The overall survival time was defined as the duration from initial symptoms onset to death.

| Statistical analysis
Data were analyzed using the SPSS version 19.0 software (SPSS Inc.,). All data were expressed as mean ± standard deviation (SD) and median (interquartile range, IQR) for normally and non-normally distributed data, respectively, and as a percentage when appropriate. To compare difference between groups, independent samples T-tests and Mann-Whitney tests were used for continuous variables that were normal and not normally distributed, respectively. The chi-squared or Fisher's exact tests were used for categorical data as appropriate. Univariate and multivariate binary logistic regression model was used to identify factors associated with adverse events.
The possible risk factors were evaluated by Kaplan-Meier survival analysis along with log-rank tests. p value <.05 (two-tailed) was considered statistically significant.

| Study population
The clinical characteristics and treatment strategies of the study patients are shown in Table 1. Of the 52 patients included, there were 19 men and 33 women. The mean age was 48.4 ± 1.8 years (range from 16 to 69 years). There were 34 cases with polymyositis and 18 cases with dermatomyositis. The chief complaint reported were dyspnea, palpitation, and skeletal muscular weakness.
All the involved patients were followed for a median of 2 years (IQR 0.8-3.0 years) and 22 deaths were recorded. Cardiogenic death was noticed in 20 cases; the other two cases of deaths were caused by infective complication and progressive respiratory failure, respectively. Twenty-six cases were readmitted due to HF and five cases were implanted with implanted cardiac device, including 3 of implantable cardioverter defibrillator. IIM-IC patients that required rehospitalization for HF or suffered from all-cause death had worse baseline functional status (greater percentage NYHA Class III-IV) and higher levels of NT-proBNP compared to IIM-IC patients who did not have adverse events (Table 1). Abbreviations: Hs-CRP, High-sensitivity C-reactive protein; NT-proBNP, N-terminal pro-brain natriuretic peptide; ACEI, Angiotensin-converting enzyme inhibitors; ARB, Angiotensin receptor blockers.

F I G U R E 1 Flow diagram of the patients included in the study
TA B L E 1 Clinical data of patients with idiopathic inflammatory myopathy and inflammatory cardiomyopathy

| Rhythm and rate
The average heart rate recorded by the electrocardiogram was 82 ± 19 beats per minute. Pause greater than 2 seconds in sinus rhythm or greater than 3 seconds in AF and advanced atrioventricular block were noticed in 11 and 2 patients, respectively. Premature ventricular contraction was present in 90.4% of involved patients, while ventricular tachycardia could be seen in more than two thirds of our cases (67.3%) (Figure 2a,b,d). Most of the patients manifested atrial arrhythmia, including paroxysmal AF (40.4%), persistent AF (25.0%), and atrial flutter (11.5%) (Figure 2d). No difference of above parameters could be observed between patients with or without adverse events (Table 2).

| Waveform
More than two thirds of the patients showed left deviation of elec-  (Table 2).

| Cardiac structural and functional features
The echocardiographic reports of all patients were consistent with the morphological feature of DCM and were mainly manifested as mildly dilated left ventricle (LVEDD: 53.7 mm ± 7.9 mm) and impaired left ventricular systolic function (LVEF: 37.8% ± 11.5%). Twenty-six patients (50.5%) met the definition of reduced LVEF. Left ventricular diastolic dysfunction was common (29/52). There were 30 patients with left ventricular enlargement, with a LVEDD of 53.7 ± 7.9 mm (Table 2). A typical example of echocardiography is shown in Figure   S1C.

| Association between cardiac parameters and adverse events
Univariate Cox regression analyses were used to explore the elec-  (Table 3).
Kaplan-Meier estimates according to electrocardiogram and echocardiography are shown in Figure 3. The results disclosed that TA B L E 2 Electrocardiographic and echocardiographic characteristics of patients with or without adverse outcome

| DISCUSS ION
Although HF is a common and potentially fatal complication of IIM, studies are limited for patients with autoimmune disease associated TA B L E 3 Cox regression analyses between all-cause death and parameters of electrocardiogram and echocardiography (N = 52)  IC. Additionally, it is crucial to distinguish these patients from those with primary DCM, as their management may differ. For instance, immunosuppressive therapies may have benefits on improving cardiac function and prognosis through reduction of active myocardial inflammation, beyond that of standard HF medications. Hence, it is crucial for cardiologists to differentiate IC from primary DCM. Here, we described the electrocardiographic and echocardiographic features of patients with IIM-IC, reporting high frequencies of Q-wave abnormality, poor R-wave progression on precordial leads, atrial fibrillation, and ventricular arrhythmia. Specifically, we found that Q-wave abnormality and reduced LVEF (≤40%) were independent predictors for adverse events and shorter survivals.
The electrocardiogram is an initial, sensitive, and noninvasive examination to identify myocardial involvement in patients with IIM. Electrocardiographic alterations were reported in 30%-80% of cases of IIM (Stern et al., 1984). Rhythm and conduction abnormalities were concluded as the most frequently reported cardiac abnormalities in IIM (Diederichsen, 2017). Especially, left anterior hemiblock and right bundle branch block were considered as the most represented alterations (Mann et al., 2014). On the other hand, although premature ventricular contraction was frequent, ventricular tachycardia had been documented rarely in PM previously (Adler et al., 2002, Lundberg, 2006. However, we found more than two thirds of our patients suffered from ventricular tachycardia, which was a risk factor of sudden death in the general population, suggesting it as a sensitive clinical marker to reflect seriously involved myocardium (Diederichsen, 2017).
We observed that the abnormal Q waves were mainly distrib-  (Rai et al., 2016).
Abnormal Q waves in adjacent leads usually indicate a prior history of myocardial infarction, and the scarring area might become an origin of ventricular arrhythmia, the latter of which could lead to sudden cardiac death. Considering that all the involved patients had be excluded from coronary artery disease through coronary angiography, the Q-wave abnormality in our patients could not be explained by "type 1 myocardial infarction." It is more accurate to describe the Q waves as an abnormal repolarization pattern, suggesting an occurrence of transmural myocardial damage in patients with IIM-IC.
Moreover, the frequency of Q-wave abnormality was associated with adverse outcome, which was reported the first time in patients with IIM, indicating that the myocardial involvement might lead to cardiogenic re-admission or cardiogenic death. Based on our analysis, electrocardiogram may help to identify IIM patients with potential myocardial involvement, who may be at higher risk for heart failure hospitalization and/or death.
Similarly, echocardiography might serve as a noninvasive and practical test for further risk stratification after initial evaluation with troponin and electrocardiogram. According to our study, left ventricular diastolic dysfunction was present in 55.8% and reduced LVEF in 50.5% of patients diagnosed with IIM-IC. Additionally, we demonstrated abnormalities in ventricular size, wall motion, pulmonary artery pressure, and pericardium in patients with IIM-IC.
The possible pathophysiology of ventricular dysfunction includes (i) myocardial inflammation leading to myocyte degeneration and interstitial fibrosis (Schwartz et al., 2016), and (ii) alteration of microcirculation, hyperplasia of intima, and sclerosis of tunica media (Mann et al., 2014). Inflammation can be determined by endomyocardial biopsy meeting the definition of myocarditis, or by CMR imaging (Mavrogeni et al., 2016) demonstrating characteristic myocardial edema and fibrosis (Pipitone, 2016). Considering the availability, safety, and cost, echocardiography may be a more feasible examination for initial assessment of IIM-IC.
In a prospective research involving 91 subjects with PM and DM, 22 cases died after a median follow-up of 8.7 years. The heart involvement (HR = 1.8) was the independent risk factors of mortality (Danieli et al., 2014). Although cardiac manifestations in IIM have been described as potentially lethal (Dilaveris et al., 2012), the epidemiology of IIM-IC and prognostic factors of poor outcome have not previously been well established. We therefore explored longitudinal follow-up of these rare cases, analyzing overall and cardiacspecific outcomes. Interestingly, we found that the adverse events were seen in more than 50% of our patients after a median follow-up of 2 years, while the proportion of cardiac death in all-cause death was as high as 90.9%. Especially in the cases with reduced LVEFs, primary and secondary endpoints were both increased significantly.
Additionally, survival time was decreased significantly in IIM with reduced LVEF, dilated left ventricle (LVEDD >60 mm), Q-wave abnormality, and low voltage of the limb leads. Since cardiac involvement is the leading cause of death in IIM, we suggest enhanced surveillance of patients with IIM-IC, particularly if Q-wave abnormality or reduced LVEF are present.
There were several limitations to this study. First, this was a single-center, retrospective study. Second, medication interventions might influence the overall outcome; however, they were not adjusted in the regression analyses as confounding factors, because of the limited sample size, complicated, and personal-specialized immunosuppressive therapies, and the fact that there was no management guideline or expert consensus on IIM-IC.

| CON CLUS ION
In our retrospective study of patients with IIM-IC, we found that Qwave abnormality and LVEF ≤40% were associated with adverse clinical events as well as all-cause mortality. We therefore recommend enhanced attention to diagnosis, management, and surveillance of these patients, including referral to cardiology, close follow-up, and consideration of immunosuppressive therapy.

CO N FLI C T O F I NTE R E S T
Prof Ligang Fang has grant support from the Chinese Academy of Medical Sciences (2019-I2M-2-003), Dr Yingxian Liu has grant support from the National Natural Science Foundation of China (82000470), and the other authors report no conflicts.

CO N S E NT FO R PU B LI C ATI O N
Not applicable.

AUTH O R CO NTR I B UTI O N S
Ligang Fang and Wei Chen performed conception and design; Ligang

E TH I C S S TATEM ENT
This study is approved by the Ethics Committee of the Peking Union Medical College Hospital and has been registered in Clinical trials (NCT 03885375).

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding authors upon reasonable request.