Management of Cardiac Tamponade During Systemic Lupus Erythematosus Flare with Significant Pericardial Effusion: A Case Report

Introduction

Large pericardial effusion (defined as greater than 500 mL) in systemic lupus erythematosus (SLE), with associated cardiac tamponade, is an uncommon but life-threatening complication of serous involvement in SLE [1]. The strongest predictor of tamponade is size of the effusion [2]. Specific risk factors include female sex, positive anti-nucleosome antibody, and lower serum C4 levels at diagnosis [3]. A case series by Maharaj et al demonstrated that the majority of cases are managed with pericardiocentesis; however, there are also reports of successful conservative management [4,5]. The size of the effusion is rarely reported in the literature and no clear guidelines exist as to when to utilize invasive treatment such as pericardiocentesis. The incidence of tamponade in SLE is not well established, although a study of 409 patients by Gosami demonstrated a 5% occurrence with approximately half of these patients having clinical signs of tamponade at diagnosis [6]. We present a case of recurrent pericarditis in a patient previously diagnosed with SLE on maintenance treatment which resulted in cardiac tamponade.

Case Report

A 38-year-old woman with a history of SLE presented to the emergency department with 3 weeks of worsening nausea, vomiting, subjective fevers, and dyspnea with associated pleuritic central chest pain and orthopnea. She denied abdominal pain, diarrhea, and joint pain. No florid neuropsychiatric features were present and there was no cutaneous involvement. On examination, there were 2 clear heart sounds without added murmur and peripheral pulses were regular and bounding. There was dullness to percussion, without lung sounds to the apex on the left. Pitting peripheral edema was present to the level of the knees. The abdomen was soft and nontender. Vital signs demonstrated a temperature of 38.6°C, oxygen saturations of 90% on 2 liters per minute supplemental oxygen via nasal prongs, blood pressure of 120/76 mmHg, heart rate of 120 beats per minute and a respiratory rate of 22 breaths per minute.

Her initial SLE diagnosis occurred at age 34 after presenting with fever, fatigue, polyarticular synovitis, and a malar rash. This was complicated by pericarditis without pericardial effusion and Class IV nephritis with induction of remission by high-dose pulsed methylprednisolone and maintenance of remission with hydroxychloroquine, mycophenolate, and corticosteroids. Remission had been relatively well maintained since initial presentation with no significant disease exacerbations despite mild persistent microscopic hematuria and proteinuria. Other past medical history was remarkable for gastro-esophageal reflux disease and rosacea. Medications at the time of admission were candesartan, cholecalciferol, hydroxychloroquine, melatonin, mycophenolate mofetil, norethisterone, prednisolone, and trimethoprim/sulfamethoxazole.

Chest X-ray was concerning for an enlarged mediastinum with rightward shift and a large left-sided pleural effusion, as can be seen in Figure 1. Laboratory testing revealed a C-Reactive Protein level of 224.5 mg/L, anti-nuclear antibody (ANA) titer >1280 (reference range, <160) in a homogenous staining pattern, and double-stranded DNA (chemiluminescent immunoassay) titer >666 (reference range, <27). The patient’s ANA titer at diagnosis was >1280 and had not been repeated. Serum albumin was 22 g/L (reference range, 35–50 g/L). Mild anemia was present, with hemoglobin of 101 g/L (reference range, 115–165 g/L) and serum lactate dehydrogenase was elevated at 352 U/L (reference range, 120–250 U/L). An anti-phospholipid syndrome screen was negative. Urine protein-creatinine ratio was 132 mg/mmol (reference range, <21 mg/mmol) without the presence of erythrocytes. Serum creatinine was raised to 111 mmol/L from a baseline of 80 mmol/L. Complement studies were low-normal (C3: 0.97 g/L, C4: 0.12 g/L); however, this was following high-dose corticosteroid administration on presentation and may have improved from initial values. Baseline complement levels taken as an outpatient 3 months prior during an asymptomatic period were similar (C3: 0.93 g/L, C4: 0.12 g/L). Systemic Lupus Erythematosus Disease Activity Index (SLEDAI) was calculated to be 29 at presentation.

The patient was initially treated with high-dose intravenous hydrocortisone, vancomycin, and ciprofloxacin out of concern for sepsis due to fever, tachycardia, and her oxygen requirement. However, antibiotics were ceased after 72 hours of negative blood cultures. Twenty-four hours after admission, the patient had a medical emergency call for an episode of tachycardia to 140 beats per minute associated with worsening retrosternal chest pain, which had initially been attributed to the large left pleural effusion. Electrocardiogram demonstrated widespread scalloped ST segment elevation without obvious electrical alternans. Systolic blood pressure at this time was 120 mmHg, from a baseline of 145 mmHg 2 hours prior, with a heart rate of 80 beats per minute. The pulse pressure was 50 mmHg. An oxygen requirement of 0.7 liters per minute had increased to 2 liters per minute to maintain oxygen saturations of 94%. New atrial fibrillation in combination with chest X-ray findings of widened mediastinum in a patient with SLE and clinical signs of fluid overload raised concern for pericardial effusion. She proceeded to transthoracic echocardiogram which demonstrated a large circumferential pericardial effusion with sonographic features of cardiac tamponade including dilated inferior vena cava with blunted inspiratory collapse. Ultrasound imaging frames of the effusion can be seen in Figures 2 and 3. The effusion measured 7.0 cm to the left ventricle in the parasternal short axis view. However, this measurement was not saved to the device. The combination of tachycardia, relative hypotension, and evidence of elevated right filling pressures on echocardiogram in the context of a massive pericardial effusion was felt to be consistent with cardiac tamponade rather than other causes of elevated right atrial filling pressures such as pulmonary hypertension. Due to the clinical deterioration of the patient with increasing oxygen requirement, worsening tachycardia, and relative hypotension, the decision was made to proceed with invasive pericardiocentesis.

Thoracocentesis was performed via ultrasound guidance for the left-sided pleural effusion which yielded 2200 mL of straw-colored fluid via a 6.3 French pigtail catheter before removal. Pericardiocentesis was then performed with ultrasound and fluoroscopy guidance with 1000 mL of haemoserous fluid being aspirated from the pericardium. The patient remained hemodynamically stable throughout the procedure. A pigtail drain was left in situ, and it drained 500 mL of fluid followed by a further 100 mL the next day, resulting in a total volume of 1600 mL drained from the pericardium. Repeat transthoracic echocardiography was performed prior to drain removal which demonstrated that only a trivial pericardial effusion remained. Pericardial fluid analysis revealed a sterile exudate consistent with pericardial effusion caused by SLE exacerbation. Specific laboratory values included glucose of 6.7 mmol/L, total protein of 40 g/L, lactate dehydrogenase of 1585 U/L, and abundant leukocytes on microscopy without other microorganisms identified. The development of cardiac tamponade in this case was felt to be due to inadequate disease control despite maintenance therapy, considering there was no history of treatment gaps or reduced adherence to therapy.

In the setting of relapsed SLE complicated by massive pericardial effusion, induction of remission was attempted via intravenous cyclophosphamide 500 mg, 2 weeks apart following the Euro-Lupus protocol, and 1 mg/kg oral prednisolone with a planned slow wean over the subsequent months [7]. Subcutaneous goserelin 3.6 mg was administered for fertility prophylaxis prior to the first dose of cyclophosphamide as an inpatient, and the second dose was administered as an outpatient following discharge. Oral furosemide 40 mg once daily was commenced for peripheral fluid overload.

Repeat echocardiography performed 2 weeks after the pericardiocentesis demonstrated no residual effusion. The patient was followed up as an outpatient, and after induction of remission with cyclophosphamide, was recommenced on mycophenolate maintenance therapy with periodic rituximab as a second maintenance agent. Her hemoglobin improved to 116 g/L at 3 weeks post-presentation without specific treatment after suppression of the underlying disease process. Six months later, her double-stranded DNA titer had fallen to 103 (reference range, <27), and serum complement levels had improved (C3: 1.17 g/L, C4: 0.23 g/L).

Discussion

Pulsed high-dose glucocorticoids remain the mainstay of managing SLE flares with pericardial involvement, with additional immunosuppression frequently required in refractory or relapsed cases [8]. Pericarditis and an associated exudative pericardial effusion are the most common cardiac complications, and arise via immune complex deposition and subsequent fibrinoid necrosis of the pericardial sac associated with perivascular infiltration by mononuclear cells [9]. Inflammation from pericarditis results in extravasation of fluid into the pericardial space which can result in pericardial tamponade, reducing end-diastolic filling volume and impairing cardiac output. These complications are generally responsive to standard pericarditis medical therapy such as non-steroidal anti-inflammatory drugs and colchicine [9]. Despite this, pericarditis has been reported to arise despite adequate control of the underlying systemic disorder, such as that described in Stone et al [10]. They describe a case of a large effusion (1300 mL) developing immediately after biochemical remission of an SLE flare in a 32-year-old woman treated with methylprednisolone and mycophenolate.

Massive effusions with features of tamponade are uncommon, with varied reports of management, ranging from pericardiocentesis to expectant treatment whilst SLE remission is achieved [4]. Table 1 outlines management options used for some of the largest effusions outlined in the current literature. Regardless of treatment, development of tamponade is a significant marker of mortality in patients with SLE [11]. The decision to proceed to interventional procedures remains at the discretion of the treating clinician. There are circumstances when pericardiocentesis may be contraindicated, such as significant anaesthetic risk or bleeding diatheses. Grenader and Shavit describe a case of SLE-associated cardiac tamponade successfully managed with intravenous immunoglobulin [12]. Belimumab, a novel B-cell activating factor (BAFF) monoclonal antibody, has been successfully used for pericarditis and cardiac tamponade in cases of refractory SLE, but its prohibitive cost generally precludes its use as a first line agent [13].

A study by Gosami et al of 409 SLE patients (12 presenting with tamponade) identified pleuritis, ANA status, and size of the effusion as risk factors for developing tamponade [6]. In the case of effusions resistant to an initial pericardiocentesis, they were managed with repeated drainage, pericardial window, or more significant immunosuppression such as cyclophosphamide. In a retrospective cohort study by Rosenbaum and colleagues of 71 patients with SLE, 41 developed a pericardial effusion, 9 demonstrated signs of tamponade, and 5 required a pericardial window [3]. They identified reduced serum C4 levels, female sex, concurrent renal disease, and hemolytic anemia as potential risk factors for development of tamponade. In our case, the presence of lupus nephritis and ongoing proteinuria were the strongest risk factors for the development of pericardial tamponade, and tighter control of the underlying disease process may have reduced the severity of her exacerbation.

Conclusions

Cardiac tamponade is an uncommon but life-threatening complication of SLE with cardiac involvement. We believe this case represents one of the largest pericardial effusions resulting in tamponade reported to date, and the case highlights the possibility of tamponade occurring at any stage of the disease course, not just as an initial presentation. Treatment of the underlying systemic disease in addition to cardiac-specific intervention is essential to the successful management of SLE with cardiac complications. Due to its rarity, there remain no guidelines on the specific management of SLE-associated cardiac tamponade. Guidance for cases where contraindications to pericardiocentesis exist, such as bleeding diatheses, which are common in SLE, requires more research.