A case of infective endocarditis mimicking antineutrophil cytoplasmic antibody-associated vasculitis

Abstract

Infective endocarditis (IE) is not a common disease, but it remains a serious condition. Antineutrophil cytoplasmic antibodies (ANCA) are often positive in IE, and discrimination between IE and ANCA-associated vasculitis is important because misdirected selection of therapy can lead to catastrophic consequences. We report a case of IE mimicking ANCA-associated vasculitis in which we were able to make a correct diagnosis and perform treatment. This case suggests that it is important to consider IE as a differential diagnosis in ANCA-positive patients.

Learning objective

Antineutrophil cytoplasmic antibodies (ANCA) are associated with primary systemic vasculitis. However, ANCA have also been described in other conditions and infective endocarditis (IE) was considered an important cause of ANCA.

Discrimination between IE and ANCA-associated vasculitis is important, although it is sometimes difficult. We report a case of IE mimicking ANCA-associated vasculitis. ANCA-positive patients with nonspecific symptoms should be suspected of having IE, checked for heart murmurs, and tested by echocardiography and blood cultures.

Introduction

Infective endocarditis (IE) is not a common disease, but it can be fatal due to a number of serious complications unless the patient is diagnosed correctly and receives appropriate therapy in the early phase. Even if treated appropriately, the in-hospital mortality rate is still high. Cases of IE in which dermatological manifestations similar to vasculitis were an initial presentation, and those in which antineutrophil cytoplasmic antibodies (ANCA) were positive have been reported [1,2]. On the other hand, vegetation is sometimes detected on the cardiac valve in patients with vasculitis, which is known as non-bacterial thrombotic endocarditis [3]. Moreover, IE and vasculitis are likely to share common symptoms, e.g. fever, malaise, and weight loss, making the differential diagnosis of IE and ANCA-associated vasculitis difficult. We report a case of IE mimicking ANCA-associated vasculitis.

Case report

A 52-year-old man started to feel a sense of fatigue one month previously. He developed leg edema and purpuric rash several days prior to visiting the department of general medicine of our hospital. His blood pressure was 155/97 mmHg, pulse rate was 107/min and regular, respiratory rate was 24/min, percutaneous oxygen saturation was 90% in room air, and body temperature was 37.9 °C. He had an enlarged spleen, pansystolic murmur over the apex, and nonpalpable purpura and pitting edema on his bilateral legs (Fig. 1A). Laboratory tests revealed the following values: leukocyte count 12.9 × 10³/mm³, with 92.2% neutrophils, 5.0% lymphocytes, 0.0% eosinophils; C-reactive protein 4.25 mg/dl; hemoglobin 9.1 g/dl; albumin 2.6 g/dl; serum creatinine 0.66 mg/dl. The procalcitonin value was 0.222 ng/ml. The plasma N-terminal pro-brain natriuretic peptide level was markedly increased up to 10,158 pg/ml. The serum IgG level was as high as 1829 mg/dl and rheumatoid factor (RF) was also as high as 135 IU/ml. The antinuclear antibody was negative and the serum complement 3 level was within the normal limit. The cytoplasmic antineutrophil cytoplasmic antibodies (C-ANCA) titer was as high as 11.2 U/ml (reference range < 3.5); however, the perinuclear ANCA (P-ANCA) titer was not high. Urinalysis revealed occult blood and protein. On his chest X-ray, the cardiothoracic ratio was 61% and bilateral costo-phrenic angles were dull. Electrocardiography demonstrated no ST abnormality. He was diagnosed with acute heart failure, IE, and ANCA- associated vasculitis, and referred to our department to evaluate cardiac function and to the dermatology department to perform skin biopsy. On transthoracic and transesophageal echocardiography, there was a mass suggesting vegetation on the mitral valve of 25*16 mm in size (Fig. 2), and the left ventricular ejection function was approximately 40%, with moderate mitral regurgitation. We strongly suspected IE, performed two sets of blood cultures, and started antibiotic therapy and diuretics. Blood cultures were positive for α hemolytic streptococcus. The pathological skin biopsy specimen exhibited leukocytoclastic vasculitis with inflammatory infiltrates, nuclear dust, and fibrinoid necrosis (Fig. 3A), but not ANCA-associated vasculitis. Immunostaining revealed the deposition of compliment C3 and IgA on the vascular walls (Fig. 3B and C). Therefore, we diagnosed him with IE and secondary leukocytoclastic vasculitis. Magnetic resonance imaging of the brain demonstrated multiple infarcts and microbleeds that were considered complications of IE. We continued antibiotic therapy for 4 weeks. The non-palpable purpura disappeared on day 5 (Fig. 1B) and his clinical status, including heart failure, improved on the same day. He was discharged and referred to the department of cardiac surgery for mitral valve replacement.

Fig. 1.

Photograph of legs of the patient. (A) On arrival. (B) After antibiotic therapy. There was nonpalpable purpura on the bilateral legs on arrival and it disappeared after antibiotic therapy.

Fig. 2.

Transthoracic and transesophageal echocardiography. There was vegetation at the point indicated by the arrow on the mitral valve.

Fig. 3.

Pathological skin biopsy specimen. (A) Leukocytoclastic vasculitis with inflammatory infiltrates, nuclear dust, and fibrinoid necrosis. (B) Immunostaining by complement C3. (C) Immunostaining by IgA. Deposition of C3 and IgA was observed on the vascular walls.

Discussion

ANCA are associated with primary systemic vasculitis, but they have also been described in other conditions such as malignancies, drug-induced vasculitis, and infection [4]. IE is also considered an important cause of ANCA. The standard medical therapy for ANCA-associated vasculitis consists of glucocorticoid and immunosuppressive agents which are counterproductive for IE. Therefore, misdirected selection of therapy can lead to catastrophic consequences and differential diagnosis between IE and ANCA-associated vasculitis is important. In previous cases, ANCA were positive in approximately 20% of IE patients and IE was considered an important cause of ANCA [4]. The characteristics of ANCA-positive IE compared with ANCA-negative IE are younger age, longer duration of symptoms before diagnosis, more frequent occurrence of echocardiographic vegetations, multiple valve involvement, and above-normal serum nonspecific inflammatory markers such as IgG or RF [4,5]. ANCA-positive IE is often diagnosed at a later stage because symptoms are likely to be non-specific (weight loss, asthenia) in the early stage, and ANCA-positive IE often exhibits a subacute course. The long duration of chronic inflammation may be associated with a high level of IgG or RF. The pathogenesis of ANCA in infection is considered to be immune dysfunction following chronic inflammation in response to microbial peptides, upregulation of autoantigen genes, and molecular mimicry between microbial antigens and autoantigens [5,6]. Our patient had the following features of ANCA-positive IE: he was relatively young (52-year-old), presented with non-specific symptoms, and had a high serum IgG level and RF.

ANCA-positive IE predominantly exhibits skin and kidney lesions, and lung and articulation lesions are less common than in ANCA-associated vasculitis. The vasculitis in ANCA-positive IE is caused by immune complex-mediated complement deposition rather than by the pauci-immune inflammation that occurs in ANCA-associated vasculitis [1,4]. The pathological findings of vasculitis associated with immune complexes are consistent with those of leukocytoclastic vasculitis with inflammatory infiltrates and fibrinoid necrosis, as in our patient. Leukocytoclastic vasculitis reportedly occurs in approximately 4% of patients with IE [7]. Therefore, our final diagnosis was IE and secondary leukocytoclastic vasculitis, not ANCA-associated vasculitis.

According to the Japanese Circulation Society Guideline, early surgery should be considered in the situations as follows: progressing heart failure, uncontrolled infection, and high risk of embolism [8]. Regarding our patient, heart failure and infection were well controlled, but there was a large vegetation, as well as cerebral microbleeds and subarachnoid hemorrhage. Therefore, we decided to continue antibiotic therapy considering the risk of early surgery after the heart team discussion. Therapeutic strategy for IE should be considered carefully throughout the entire duration of therapy.

Conclusion

We report a case of IE mimicking ANCA-associated vasculitis. The differential diagnosis between IE and ANCA-associated vasculitis is important because misdiagnosis leads to inappropriate therapy. ANCA-positive patients with nonspecific symptoms and skin lesions suggesting vasculitis should be suspected of having IE, checked for heart murmurs, and tested by echocardiography and blood cultures.

Declaration of competing interest

All authors have no conflicts of interest to disclose.