Fatal Cardiomyopathy Secondary to Seronegative Immune-Mediated Necrotizing Myopathy: A Case Report

Introduction

Idiopathic inflammatory myopathies (IIM) are autoimmune disorders that cause inflammation of muscle tissue and necrosis of muscle fibers. IIM primarily affects proximal muscles groups, including the deltoid and psoas, although involvement of neck flexor, pharyngeal, and respiratory muscles can lead to dysphagia and dyspnea [1]. Clinically, patients generally present with symmetric proximal upper and lower extremity weakness. IIM can be classified into subgroups of polymyositis, dermatomyositis, inclusion body myositis, and immune-mediated necrotizing myopathy (IMNM). IMNM can be further subdivided based on presence of antibodies against either signal recognition particle (SRP), hydroxy-3-methylglutaryl-CoA reductase (HMGCR), or neither (seronegative) [2,3]. Seronegative IMNM is less common, occurring in approximately 10% to 12% of cases [3]. Seronegative IMNM is more frequently associated with an underlying malignancy, with malignancy identified in approximately 20% of cases [2].

Cardiac involvement in IIM is thought to be uncommon, and the prevalence of cardiac involvement in IIM likely varies. For example, 9.3% of polymyositis and 4.7% of dermatomyositis cases demonstrated symptoms of heart failure or myocarditis, and/or abnormalities in electrocardiography and echocardiography [4]. Cardiac involvement in IIM may depend on serology. In patients with polymyositis/dermatomyositis, strongly positive anti-mitochondrial antibodies (AMA) were associated with severe cardiac disease [5]. Cardiomyopathy has been described in case reports of SRP-positive IMNM [6,7], and estimation of cardiac involvement varies greatly from 2% of cases to almost 50% of cases [8,9].

Cardiac involvement in seronegative IMNM is not well understood. We present a rare case of a patient who had treatment-refractory myocarditis secondary to seronegative IMNM leading to progressive cardiomyopathy and death. We discuss this case in the context of the few published cases in the literature and review the limited available data on treatment and prognosis.

Case Report

A 35-year-old previously healthy White man with minimal past medical history of obesity and untreated hypertriglyceridemia presented with progressive proximal upper and lower extremity muscle weakness, myalgias, Raynaud phenomenon, and unintentional weight loss. Physical examination demonstrated mild weakness in the proximal muscles of the bilateral upper and lower extremities. The examination was otherwise negative for skin changes, synovitis, sclerodactyly, active Raynaud phenomenon, or digital ulceration. Family history was pertinent for rheumatoid arthritis in his mother and his father. Relevant social history included tobacco use (5 cigarettes daily for 8 years but quit 3 years prior to presentation), and no significant alcohol use or other substance use. On presentation, he was found to have an elevated creatinine kinase level of 7500 (reference range, <308 U/L), C-reactive protein level of 37 (reference range, <5mg/L), and ferritin level of 740 (reference range, <275 ng/mL). Initial differential diagnosis included idiopathic inflammatory myopathy, metabolic myopathies, scleroderma, and overlap syndrome, given the patient’s history of Raynaud syndrome, muscular dystrophy, or a paraneoplastic process. Magnetic resonance imaging (MRI) demonstrated diffuse signal abnormality and enhancement of the left pelvis and thigh, most prominent in the rectus femoris. Autoimmune serologies, including antinuclear antibody (ANA), Sjögren syndrome antigen A (SSA), Sjögren syndrome antigen B (SSB), anti–double-stranded DNA, rheumatoid factor, cyclic citrullinated peptide, anti-HMGCR, anti-SRP, and an additional myositis panel for myositis-specific or myositis-associated antibodies (Table 1), were all negative. Muscle biopsy of the left rectus femoris demonstrated numerous muscle fibers at varying stages of necrosis and regenerating fibers, suggesting active myopathy (Figure 1). In trichrome-stained sections, ragged red fibers were not observed. Cytochrome C oxidase–negative fibers were not seen, beyond staining of necrotic fibers. Macrophages were identified in acid-phosphotase–stained sections (Figure 1C). Immunostaining for lymphocytes was not performed as inflammatory reaction was not observed in hematoxylin and eosin (H&E)–stained sections. Major histocompatibility complex class I expression was observed (Figure 1D), as was membrane attack complex deposition (Figure 1E). Seronegative immune-mediated necrotizing myopathy was diagnosed. The patient was treated with oral prednisone 80 mg daily for 2 weeks, followed by a 10 mg taper every 2 weeks, with minimal improvement in symptoms. A workup for underlying malignancy, including computed tomography of the chest and abdomen, MRI of the brain, and colonoscopy, was negative. Due to ongoing symptoms of weakness on prednisone, treatment was escalated to intravenous (IV) methylprednisolone 1 g weekly for 8 weeks, oral methotrexate 12.5 mg weekly with planned up-titration, and intravenous immunoglobulin (IVIG) 0.4 g/kg/day for 5 total days given monthly. Despite this, the patient had persistent creatinine kinase elevation over 1500 U/L and no clinical response to treatment.

Seven months after the initial presentation, the patient re-presented with the acute onset of heart failure. An echocardiogram demonstrated reduced biventricular systolic function and was markedly changed from the echocardiogram obtained at initial diagnosis, which had not demonstrated any significant pathology. Coronary angiography showed no ischemic disease. Cardiac MRI demonstrated cardiomyopathy, with left ventricular ejection fraction of 30%, moderately reduced right ventricular systolic function, and subendocardial hyperenhancement, findings suggestive of myocarditis. He reported no infectious symptoms or viral prodrome. IVIG was discontinued given suspected disease progression, and he received induction therapy with IV rituximab 1000 mg with a repeat dose 14 days later, along with oral dexamethasone 8 mg daily with a planned taper by 1 mg per week. Two months later, he developed recurrent episodes of ventricular tachycardia, requiring the placement of an implantable cardioverter-defibrillator. Repeat cardiac MRI demonstrated ongoing myocarditis despite treatment. Given the lack of disease response, methotrexate was discontinued, and the patient was transitioned to oral mycophenolate 1500 mg twice daily, with dexamethasone increased back to 8 mg daily.

A repeat serologic workup demonstrated positive ANA at 1: 160 (speckled), and borderline positive AMA (M2) antibody 0.2 U by enzyme immunoassay (reference range, <0.1 U, negative; 0.1–0.3 U, borderline). Further evaluation of AMA by another method was not pursued, as this was below the threshold for weak positivity. Repeat serologic panel, including SSA, SSB, RNP, Smith, Scl-70, anti-centromere, anti–RNA polymerase III, anti-HMGCR, anti-SRP, and a repeat myositis panel as described above, additionally including anti-SAE1 (Table 1), β2-glycoprotein I, and anti-cardiolipin antibodies, was negative. Given the rarity of advanced cardiomyopathy secondary to IMNM, alternative diagnoses were reconsidered including, infectious myocarditis, cardiac sarcoidosis, and giant cell myocarditis. A workup for infectious causes of myocarditis was negative, including HIV types 1 and 2 antigen and antibody screening, plasma Enterovirus polymerase chain reaction (PCR), plasma parvovirus B19 PCR, plasma cytomegalovirus real-time PCR (RT-PCR), plasma Epstein-Barr virus RT-PCR, blood Tropheryma whipplei PCR, Borrelia burgdorferi immunoglobulin M and immunoglobulin G, and Trypanosoma cruzi antibody. Prior testing for serum hepatitis B virus surface antigen and core total antibody was negative, as was hepatitis C virus antibody. A Karius Spectrum (Karius, Inc, Redwood City, CA, USA) blood test for cell-free DNA did not detect any pathogen at statistically significant levels. The patient did not have respiratory symptoms, and PCR studies from a nasopharyngeal swab did not detect adenovirus, influenza A or B, or SARS-CoV-2. A cardiac positron emission tomography (PET) scan was performed, demonstrating areas of perfusion defect with associated focal fluorodeoxyglucose (FDG) uptake in the anterior, anteroseptal, and anterolateral segments, suggestive of an active inflammatory myocardial process. A whole-body PET-CT was subsequently performed, demonstrating mild proximal muscle FDG activity without evidence of malignancy or extracardiac sarcoidosis.

The combined findings of the PET scans suggested minimal inflammation of the proximal muscles with active myocardial inflammation, leading to the decision to proceed with a cardiac biopsy. Endomyocardial biopsy was performed; 3 specimens of the right ventricle were taken and evaluated at over 50 levels to reduce the likelihood of tissue sampling error. Biopsy demonstrated multifocal areas of loose interstitial fibrosis (pericellular and replacement type) with fibroblasts, macrophages, and rare scattered pigment, along with myocyte hypertrophy (Figure 2). Additional stains, including immunohistochemistry, were not performed. Immunostaining for lymphocytes was not performed, as no lymphocytic inflammation was seen on H&E-stained sections. Biopsy did not demonstrate granulomatous inflammation to suggest cardiac sarcoidosis or multinucleated giant cells to suggest giant cell myocarditis. Viral PCR was not performed on biopsy tissue, as this is not standard at our institution, no viral cytopathic effect or inflammation was seen on biopsy, and the patient had a prior negative serologic workup for infection. Biopsy findings were interpreted as subacute and nonspecific; however, in the clinical context, they were favored to represent myocarditis secondary to systemic myopathy.

While hospitalized, the patient continued to experience intermittent episodes of ventricular fibrillation despite aggressive medical management. Due to the severity of disease, 10 mg/kg IV cyclophosphamide was given in addition to 0.4 g/kg/day IVIG for 5 days, oral prednisone 40 mg daily, and oral mycophenolate 1500 mg twice daily. Shortly after starting IV cyclophosphamide, he developed cardiac arrest with pulseless electrical activity; after the return of spontaneous circulation, a microaxial flow pump (Impella) was placed for circulatory support. He developed progressive cardiogenic shock with acute respiratory distress syndrome, requiring vasopressor support and intubation. After discussion with his family, he was compassionately extubated and died.

An autopsy was performed. Cardiac examination revealed severe cardiomegaly, and histology demonstrated diffuse fibrosis of the right ventricle and severe pericellular and replacement fibrosis of the left ventricle with persistent mild chronic inflammation with mononuclear cells (Figure 3A–3C). Samples of skeletal muscle and cardiac tissue were obtained and prepared with H&E stain. Histologic evaluation of the psoas re-demonstrated focal degenerating necrotic fibers with fibrosis (Figure 3D, 3E). There were no additional stains performed on autopsy tissue. There was no evidence of malignancy on autopsy. Cause of death was deemed systemic immune-mediated necrotizing myositis and myocarditis. A timeline overview of the complete clinical course is summarized in Figure 4.

Discussion

We present a fatal case of progressive, treatment-refractory cardiomyopathy secondary to seronegative IMNM. IMNM is divided into 3 subtypes based on the presence of anti-SRP antibodies, anti-HMGCR antibodies, or neither (seronegative), as in this case [2,3]. It is hypothesized that the degree of cardiac involvement may depend on IMNM subtype, although reports on the frequency of cardiac involvement in SRP-positive IMNM, HMGCR-positive IMNM, or seronegative IMNM are highly variable [8–11]. While there is a strong association between seronegative IMNM and malignancy [2], there was no underlying malignancy in this case. Despite extensive testing for additional myositis-specific or myositis-associated antibodies (Table 1), evaluation was notable only for a weakly positive ANA and a borderline positive AMA (M2 antibody), both of which were identified late in the disease course. Identification of positive AMA antibody is critical as this been linked with severe cardiac manifestations [5]. However, the borderline value here is below the level considered for weak positivity, and prior cases of AMA-associated myopathy reported by Albayda et al were characterized by strongly positive AMA. Furthermore, these patients were described as presenting clinically with mild muscle weakness initially and a chronic clinical course. This is inconsistent with the rapid progression in this case, and we interpret the borderline AMA antibody as an incidental finding in a case that is better classified as seronegative IMNM.

Life-threatening cardiomyopathy secondary to seronegative IMNM is a rare phenomenon with significant clinical importance. The strengths of the present case include the abundance of diagnostic data, multiple biopsies, advanced imaging modalities, and autopsy. The primary limitation is the difficulty interpreting the clinical significance of the findings described in this unique case. Therefore, we attempt to interpret our case in the context of the limited available literature on cardiomyopathy secondary to seronegative IMNM.

In all patients with IIM, cardiac involvement is a significant prognostic factor for death, and in polymyositis/dermatomyositis, it is the main prognostic factor for death [4]. While there is not specific data for IMNM, cardiac involvement is similarly presumed to convey a poor prognosis. Including our case report, 3 out of 4 individual case reports of patients with seronegative IMNM and clinically significant cardiac involvement led to patient death [12–15] (Table 2). Similarities from these cases include that all 4 patients were between the ages of 35 and 45 years and had minimal past medical history (with the notable exception of previous hepatitis B in Sanchez et al), and none of these patients were found to have malignancy. Additionally, these cases all had early cardiac involvement and a rapidly progressive clinical course. Myocardial histology between these cases is somewhat variable; Tsang et al demonstrated lymphocytic and monocytic myocarditis, Tamura et al demonstrated the presence of CD68-positive macrophages, and Sanchez et al demonstrated extensive fibrosis without other specific findings. We describe the presence of macrophages on the background of fibrosis seen on endomyocardial biopsy, which progressed to extensive fibrosis seen on autopsy. It is unclear why the patient described in the report by Tsang et al is unique in that they had a complete recovery, especially given that treatment interventions were similar to those of other cases. We speculate that the presence of lymphocytic and monocytic cells on biopsy may have been favorable for a response to immunosuppressive therapy; this contrasts with the remaining cases in which there is predominant fibrosis and no inflammatory response to target with immunosuppressive therapies. We further hypothesize that this represents a more acute inflammatory phase of myocarditis, which would suggest that early identification and treatment of these patients is critical before disease progression and fibrosis/remodeling occur.

Variable cardiac involvement in patients with seronegative IMNM has been reported. In contrast to these individual case reports, small case series demonstrated that subclinical cardiac involvement can be found in 50% to 70% of patients with seronegative IMNM when they are assessed by electrocardiography, echocardiography, or cardiac MRI [9–11]. Triplett et al reported on cardiac involvement in IMNM, including a subset of seronegative patients, and found that left ventricular diastolic dysfunction was the most common cardiac abnormality in their cohort of patients with seronegative IMNM [10]. They also found that of all patients with IMNM (not specifically seronegative) that had an echocardiogram repeated after treatment, several had improvement in left ventricular diastolic function. Despite the subclinical cardiac involvement described in Ma et al, these patients with seronegative IMNM had a universally adequate response to initial treatment, and only 1 patient went on to develop clinical cardiac involvement [11]. Given the prevalence of subclinical cardiac involvement described in these studies, it is unclear why so few patients progress to clinically significant cardiac disease, or how this progression occurs. To improve clinician awareness, future investigative studies should focus on identifying risk factors or biomarkers predicting cardiac involvement. While a variety of treatments have been attempted, as described in Table 2, it is not clear if any of these are effective at treating clinically significant cardiac manifestations of seronegative IMNM. This should also be a focus of future research. Interestingly, our case and that of Sanchez et al describe worsening cardiac disease with simultaneous improvements in skeletal myopathy [15]. This leads us to hypothesize that advanced cardiomyopathy secondary to IMNM requires targeted treatment distinct from treatment of skeletal myopathy.

Conclusions

In summary, subclinical cardiac involvement in seronegative IMNM may be more prevalent than previously thought. Clinically significant cardiac manifestations of seronegative IMNM are rare; however, when they occur, they likely convey a poor prognosis and are often fatal, as we have shown here. There is currently a lack of understanding of how subclinical cardiac manifestations progress, and which treatments could either prevent this progression or treat advanced cardiac disease. These should all be areas of future investigation.