Giant cell myocarditis with central diabetes insipidus: A case report

Open ArchivePublished:September 10, 2019DOI:https://doi.org/10.1016/j.jccase.2019.08.011

      Abstract

      A 51-year-old male, previously diagnosed with central diabetes insipidus due to lymphocytic hypophysitis, presented with fever and dyspnea for 1 week. On arrival, he exhibited hypotension (85/60 mmHg) and sinus tachycardia (110 bpm). His electrocardiogram revealed mild ST elevation on V2–V4. Echocardiography indicated a near-normal (50%) left ventricular ejection fraction (LVEF), although the inferior wall of the left ventricle exhibited severe hypokinesis. Fulminant myocarditis and circulatory insufficiency were suspected, and treatment with dobutamine, 3 μg/kg/min, was started. His LVEF gradually decreased to 20%. On day 17, he developed cardiogenic shock due to ventricular tachycardia and underwent peripheral venous–arterial extracorporeal membrane oxygenation and intra-aortic balloon pumping. Although he did not exhibit polyuria, intravenous vasopressin infusion (0.5 U/h) was performed to maintain normonatremia. Endomyocardial biopsy results revealed the infiltration of scattered giant cells (GCs) and extensive lymphocytes. Despite immunosuppressive therapy (methylprednisolone and cyclosporine), his cardiac function did not recover. On day 36, he received a biventricular assist device; however, he died on day 47 due to the progression of sepsis and multiple organ failure. We speculate that a deficient expression of programmed cell death protein-1 was the cause of both GC myocarditis and lymphocytic hypophysitis.
      <Learning objective: We describe a fatal case of fulminant giant cell myocarditis complicated by central diabetes insipidus due to lymphocytic hypophysitis. Normonatremia was maintained with intravenous vasopressin 0.5 U/h, and circulatory status was maintained with mechanical circulatory support. We speculate that T-cell programmed cell death protein 1 dysregulation was the common cause of the two disorders.>

      Keywords

      Introduction

      Giant cell myocarditis (GCM) is rare, but it often presents with a fulminant course and has a fatal outcome [
      • Cooper Jr., L.T.
      Giant cell myocarditis: diagnosis and treatment.
      ]. Fulminant GCM is marked by lethal arrhythmias, acute heart failure, and cardiogenic shock. The cause of GCM is not clear, but studies have suggested that the condition is mediated by T-lymphocytes [
      • Litovsky S.H.
      • Burke A.P.
      • Virmani R.
      Giant cell myocarditis: an entity distinct from sarcoidosis characterized by multiphasic myocyte destruction by cytotoxic T cells and histiocytic giant cells.
      ]. GCM is typically virus negative, and up to 20% of patients with GCM have other inflammatory or autoimmune comorbidities [
      • Cooper Jr., L.T.
      Giant cell myocarditis: diagnosis and treatment.
      ].
      Central diabetes insipidus (DI) is characterized by polyuria and polydipsia due to vasopressin deficiency [
      • Arima H.
      • Azuma Y.
      • Morishita Y.
      • Hagiwara D.
      Central diabetes insipidus.
      ]. It is known that autoimmune mechanisms contribute to the development of lymphocytic hypophysitis, which leads to central DI. Lymphocytic hypophysitis can be detected using head magnetic resonance imaging (MRI) [
      • Imura H.
      • Nakao K.
      • Akira S.
      • Yoshihiro O.
      • Takehiro S.
      • Ichiro F.
      • et al.
      Lymphocytic infundibuloneurohypophysitis as a cause of central diabetes insipidus.
      ], although a definitive diagnosis requires histologic findings of lymphocytic infiltration into the pituitary gland.
      Here we present a rare case of a patient with GCM with comorbid central DI due to lymphocytic hypophysitis.

      Case report

      A 51-year-old male visited a local hospital with a 1-week history of fever and dyspnea. Two years previously, he was diagnosed with central DI. This is because he complained of polyuria, and presented with typical lymphocytic hypophysitis observed on head MRI (Fig. 1). Therefore, he was taking oral desmopressin acetate hydrate (180 μg/day).
      Fig. 1
      Fig. 1Head magnetic resonance T1-weighted image showing a lack of hyperintense signal in the pituitary posterior lobe (arrow), thickening of the pituitary stalk (A), and enhanced intensity of the pituitary with gadolinium administration (B).
      On arrival at the hospital, he presented with a clear sensorium, hypotension [blood pressure (BP) 85/60 mmHg], sinus tachycardia (heart rate 110 bpm), 96% oxygen saturation on room air, and body temperature of 37.2 °C. No jugular venous distention, chest murmurs, rales, extremity edema, or skin rash was detected on physical examination. Laboratory examination revealed elevated aspartate aminotransferase (58 U/L), alanine aminotransferase (109 U/L), lactate dehydrogenase (239 U/L), C-reactive protein (CRP) (7.9 mg/dL), and brain natriuretic peptide (BNP) (534 pg/mL) levels. Creatine kinase levels were normal (118 U/L), but there was a slight increase in troponin I levels (2359 pg/mL; upper limit 27 pg/mL). Viral serology showed no specific findings.
      Twelve-lead electrocardiography revealed mild ST elevation on V2–V4. Echocardiography indicated a near-normal (50%) left ventricular ejection fraction (LVEF), although the apex and inferior wall of the left ventricle exhibited severe hypokinesis, with diffuse edema of the left ventricle. Coronary angiography showed his coronary arteries to be normal. The patient refused an endomyocardial biopsy; thus, histological findings could not be obtained. However, myocarditis was suspected and circulatory insufficiency was diagnosed.
      After admission, the patient was treated with dobutamine, 3 μg/kg/min, and oral desmopressin for DI was continued. He did not require mechanical circulatory support, and was not administered immunosuppression therapy including steroid or immunoglobulin. On day 9, his troponin I had not improved (2599 pg/mL) although his CRP and BNP level decreased to 1.25 mg/dL and 167 pg/mL, respectively. In terms of the echocardiogram findings, his LVEF improved to 62% with mild hypokinesis in the inferior wall. He was discontinued dobutamine administration. His dyspnea diminished, and he returned to being normothermic (36.5–36.8 °C). On day 14, fever recurred (body temperature up to 39.2 °C), although the blood culture results were negative. A follow-up 12-lead electrocardiography revealed marked ST-T abnormality and prolonged QRS width (Fig. 2A). In addition, echocardiography showed that his LVEF had decreased to 20%. On day 17, he developed cardiogenic shock due to ventricular tachycardia. He underwent peripheral venous–arterial extracorporeal membrane oxygenation (VA-ECMO) and intra-aortic balloon pumping (IABP), followed by endomyocardial biopsy. Two days later, on day 19, immunosuppressive therapy with methylprednisolone was initiated (1 g/day for 3 days and 60 mg/day thereafter), and the patient was transferred to our hospital for the consideration of heart transplant.
      Fig. 2
      Fig. 2(A) The patient’s electrocardiogram on day 14 showing marked ST-T abnormality in all leads and prolonged QRS width. (B) Echocardiography on day 19 showing the patient’s left ventricular ejection fraction to be <5%, with an edematous wall and pericardial effusion.
      At that time, with 3.5 L/min VA-ECMO and IABP support, his BP was 98/37 mmHg. We converted IABP to an Impella 2.5® peripheral left ventricular assist device (p-LVAD) (Abiomed, Danvers, MA, USA). Under mechanical circulatory support with 3.3 L/min VA-ECMO and 0.4 L/min p-LVAD, his BP was 75/70 mmHg, central venous pressure 12 mmHg, and pulmonary artery pressure 13/12 mmHg. Echocardiography revealed that LVEF had decreased to <5%, with mild pericardial effusion (Fig. 2B). When the VA-ECMO flow was reduced, this resulted in a decrease in BP with no increase in p-LVAD flow, indicating severe right-side heart failure. Without the administration of vasopressin, his urine output was 100 cm3/h. He gradually exhibited hypernatremia (159 mEq/L); thus, we initiated vasopressin, 0.5 U/h, with the concomitant use of furosemide, 40 mg/h, to maintain his urine output at 80–120 mL/h with normonatremia. Endomyocardial biopsy results (received on day 23) revealed the presence of GC infiltration with extensive CD 3 and CD 8 positive lymphocyte infiltration, interstitial edema, and myocardial necrosis (Fig. 3A, B and Supplementary Fig. 1). Therefore, the patient was diagnosed with GCM and administered intravenous cyclosporine (nadir level, 150 ng/mL). However, his cardiac function did not normalize. On day 36, he received a biventricular assist device. We obtained a resected specimen of the left ventricular apex, which showed persistent GC and lymphocytic inflammation (Fig. 3C). His condition deteriorated with multiple organ failure due to sepsis, and he died on day 47. His family did not agree to perform autopsy.
      Fig. 3
      Fig. 3Histological findings and immunohistochemistry (scale bar: 50 μm). (A, B) Endomyocardial biopsy findings on day 17 showed dense lymphocytic infiltration with scattered giant cells (inset: a giant cell) (A), and eosinophilic infiltration (B). (C) A re-examination biopsy on day 36 showed persistent inflammation. (D) Immunohistochemistry on day 23 showed a deficient expression of programmed cell death protein 1 (PD-1). The inset shows PD-1–positive T-cells of a different patient with lymphocytic myocarditis. Anti PD-1 antibody: Mouse monoclonal [NAT] to PD1 (ab52587).

      Discussion

      We report a rare case of fulminant GCM complicated by central DI. A definitive diagnosis of GCM requires the confirmation of specific histopathological findings. Unfortunately, the patient refused an endomyocardial biopsy at the time of admission, which resulted in a delay in the diagnosis of GCM. Up to 20% of patients with GCM exhibit autoimmune abnormalities [
      • Cooper Jr., L.T.
      Giant cell myocarditis: diagnosis and treatment.
      ], and the present patient had a history of lymphocytic hypophysitis, which is known to have an association with autoimmune disease [
      • Di Iorgi N.
      • Napoli F.
      • Allegri A.E.
      • Olivieri I.
      • Bertelli E.
      • Gallizia A.
      • et al.
      Diabetes insipidus—diagnosis and management.
      ]. Immunosuppressive therapy is recommended for patients with non-cardiac autoimmune diseases when an autoimmune form of myocarditis is confirmed [
      • Caforio A.L.
      • Pankuweit S.
      • Arbustini E.
      • Basso C.
      • Gimeno-Blanes J.
      • Felix S.B.
      • et al.
      Current state of knowledge on aetiology, diagnosis, management, and therapy of myocarditis: a position statement of the European Society of Cardiology Working Group on Myocardial and Pericardial Diseases.
      ]. Thus, we should have performed a more aggressive histological diagnosis during the early phase of myocarditis to detect its type and to decide whether immunosuppressive therapy was indicated.
      The incidences of GCM and central DI have been reported as 0.007% [
      • Okada R.
      • Wakafuji S.
      Myocarditis in autopsy.
      ] and 0.0004% [
      • Di Iorgi N.
      • Napoli F.
      • Allegri A.E.
      • Olivieri I.
      • Bertelli E.
      • Gallizia A.
      • et al.
      Diabetes insipidus—diagnosis and management.
      ], respectively. It is, therefore, highly unlikely that these two diseases developed independently, with separate underlying mechanisms, so we explored the possibility of a common underlying immunological abnormality.
      A literature search showed no results for myocarditis with hypophysitis. Although sarcoidosis or other granulomatous diseases can cause myocarditis and hypophysitis, there were no specific findings including elevation of serum angiotensin-converting enzyme, IgG4, and soluble interleukin 2R, enlargement of lymph node on computed tomography, and dermatologic and ophthalmologic examination.
      Anti-programmed cell death protein 1 (PD-1) antibody treatment has been reported to be a cause of central DI and myocarditis [
      • Postow M.A.
      • Sidlow R.
      • Hellmann M.D.
      Immune-related adverse events associated with immune checkpoint blockade.
      ]. In the present case, these diseases developed at different times. In addition, the recurrence of myocarditis was experienced on day 14. This unique clinical course may be explained by the knowledge of anti PD-1 antibody treatment. It is known that anti PD-1 antibody treatment has inflammatory side effects, and it affects various organs at anytime, including after cessation of the treatment [
      • Postow M.A.
      • Sidlow R.
      • Hellmann M.D.
      Immune-related adverse events associated with immune checkpoint blockade.
      ].
      Therefore, we examined the lymphocyte expression of PD-1, although the patient had no history of anti-PD-1 antibody treatment. We found that the patient exhibited low expression of T-cell PD-1 (Fig. 3D). It is known that PD-1 expression is activated by the infectious status [
      • Ishida Y.
      • Agata Y.
      • Shibahara K.
      • Honjo T.
      Induced expression of PD-1, a novel member of the immunoglobulin gene superfamily, upon programmed cell death.
      ]. However, in principle, it is difficult to quantify the expression of PD-1 by immune staining because the sensitivity and specificity of immune staining is not established. Therefore, we also examined the lymphocyte expression of PD-1 in a 22-year-old female with fulminant lymphocytic myocarditis as a positive control. (Fig. 3D inset and Supplementary Fig. 2). She presented less than 5% left ventricular ejection fraction (LVEF) at admission, and required venoarterial extracorporeal membrane oxygenation. She successfully recovered without impaired left ventricular contractility with LVEF 60% at day 14. We counted the PD-1 positive (PD-1+) and CD3 positive (CD3+) lymphocytes in 10 high-power fields (HPF). There were 78 PD-1+ in 164 CD3+ (PD-1+/ CD3+, 47.6%) in the lymphocytic myocarditis. On the other hand, there was only one PD-1+ in 253 CD3+ (PD-1+/ CD3+, 0.4%) in the present case. Therefore, we thought that low expression of PD-1 in the present case is an abnormal finding.
      PD-1 is an immune checkpoint marker that is crucial for maintaining self-tolerance [
      • Ishida Y.
      • Agata Y.
      • Shibahara K.
      • Honjo T.
      Induced expression of PD-1, a novel member of the immunoglobulin gene superfamily, upon programmed cell death.
      ]. For instance, PD-1-knockout mice have been reported to develop proliferative lupus-like arthritis and glomerulonephritis [
      • Nishimura H.
      • Nose M.
      • Hiai H.
      • Minato N.
      • Honjo T.
      Development of lupus-like autoimmune diseases by disruption of the PD-1 gene encoding an ITIM motif-carrying immunoreceptor.
      ], and anti-PD-1 antibodies can induce immune-related adverse events that involve not only the endocrine glands but also the cardiovascular system [
      • Postow M.A.
      • Sidlow R.
      • Hellmann M.D.
      Immune-related adverse events associated with immune checkpoint blockade.
      ]. We speculate that this patient had an abnormal expression of T-cell PD-1, which ultimately led to both GCM and central DI.

      Conflict of interest

      The authors declare that there is no conflict of interest.

      Appendix A. Supplementary data

      The following are Supplementary data to this article:
      • Fig. S2

        Endomyocardial biopsy findings of the present case (a–c) and control case (d–f). (a, d: CD3 staining; b,c and e,f: programmed cell death protein 1 staining). The control case was that of a 22-year-old female with lymphocytic fulminant myocarditis who successfully recovered without impaired left ventricular contractility.

      References

        • Cooper Jr., L.T.
        Giant cell myocarditis: diagnosis and treatment.
        Herz. 2000; 25: 291-298
        • Litovsky S.H.
        • Burke A.P.
        • Virmani R.
        Giant cell myocarditis: an entity distinct from sarcoidosis characterized by multiphasic myocyte destruction by cytotoxic T cells and histiocytic giant cells.
        Mod Pathol. 1996; 9: 1126-1134
        • Arima H.
        • Azuma Y.
        • Morishita Y.
        • Hagiwara D.
        Central diabetes insipidus.
        Nagoya J Med Sci. 2016; 78: 349-358
        • Imura H.
        • Nakao K.
        • Akira S.
        • Yoshihiro O.
        • Takehiro S.
        • Ichiro F.
        • et al.
        Lymphocytic infundibuloneurohypophysitis as a cause of central diabetes insipidus.
        N Engl J Med. 1993; 329: 683-689
        • Di Iorgi N.
        • Napoli F.
        • Allegri A.E.
        • Olivieri I.
        • Bertelli E.
        • Gallizia A.
        • et al.
        Diabetes insipidus—diagnosis and management.
        Horm Res Paediatr. 2012; 77: 69-84
        • Caforio A.L.
        • Pankuweit S.
        • Arbustini E.
        • Basso C.
        • Gimeno-Blanes J.
        • Felix S.B.
        • et al.
        Current state of knowledge on aetiology, diagnosis, management, and therapy of myocarditis: a position statement of the European Society of Cardiology Working Group on Myocardial and Pericardial Diseases.
        Eur Heart J. 2013; 34 (48a-48d): 2636-2648
        • Okada R.
        • Wakafuji S.
        Myocarditis in autopsy.
        Heart Vessels Suppl. 1985; 1: 23-29
        • Postow M.A.
        • Sidlow R.
        • Hellmann M.D.
        Immune-related adverse events associated with immune checkpoint blockade.
        N Engl J Med. 2018; 378: 158-168
        • Ishida Y.
        • Agata Y.
        • Shibahara K.
        • Honjo T.
        Induced expression of PD-1, a novel member of the immunoglobulin gene superfamily, upon programmed cell death.
        EMBO J. 1992; 11: 3887-3895
        • Nishimura H.
        • Nose M.
        • Hiai H.
        • Minato N.
        • Honjo T.
        Development of lupus-like autoimmune diseases by disruption of the PD-1 gene encoding an ITIM motif-carrying immunoreceptor.
        Immunity. 1999; 11: 141-151