A 10-year-old male with a past medical history of premature pubarche, mild persistent asthma, and eczema presented to the emergency department with progressive dyspnea and chest pain. On examination, he was found to be tachycardic and tachypneic. Chest radiograph demonstrated cardiomegaly, bilateral pleural effusions, and scattered atelectasis. Echocardiogram revealed a large pericardial effusion with right atrial collapse. The patient was admitted to the pediatric ICU for pericardiocentesis and drain placement. As he later became hypertensive and febrile, we will discuss how our patient’s hospital course guided our differential diagnosis and how we arrived at a definitive diagnosis using a multidisciplinary approach.
Case Description (Melanie Rosado, Medical Student) (Jordan Hall, MD, Pediatric Resident, Moderator)
The patient is a 10-year-old male with a past medical history of premature pubarche, mild intermittent asthma, and eczema who presented to the emergency department (ED) with a 4-day history of dyspnea and chest pain. His mother initially attributed his symptoms to an asthma exacerbation and treated him with inhaled albuterol. Despite this intervention, his dyspnea progressed from being exertional in nature to constant. His associated chest pain was sharp, located in the center of his chest, and worse with inspiration. The night before presentation he was unable to lie supine because of the dyspnea. He also complained of a dry cough and decreased appetite over the last 2 weeks. There was no recent history of fever or illness. Review of systems was positive for a 15 to 20 pound weight loss over the last 6 months, which his mother attributed to diet and exercise. The patient’s mother denied any medication use aside from the oral montelukast and albuterol inhaler prescribed for his asthma.
On physical examination, he was alert and interactive but visibly anxious. He was tachycardic with a heart rate of 144 beats per minute (reference range for age: 75–118), normotensive with a blood pressure (BP) of 107 of 61 mmHg, tachypneic with a respiratory rate of 32 breaths per minute (reference range for age: 18–25), and was saturating 98% in room air. He was afebrile with a temperature of 97.9 degrees F (Fig 1). His weight was 54.7 kg (98th percentile) and his height was 142 cm (70th percentile). BMI was in the 98th percentile. Head, neck, and oropharyngeal examination demonstrated dry cracked lips and pale conjunctiva. There was no cervical lymphadenopathy or jugular venous distension present. He had nasal flaring, subcostal, and intercostal retractions. Lung auscultation revealed decreased breath sounds in the bases with dullness to percussion in the bilateral lower lung fields. There was no wheezing, rhonchi, or rales present. Cardiac examination revealed distant, but otherwise normal, S1 and S2 heart sounds without any murmur, gallop, or rub. His abdomen was soft and nontender without evidence of masses or hepatosplenomegaly. Distal pulses were normal and there was no evidence of peripheral edema or clubbing. His skin, musculoskeletal, and neurologic examinations were normal.
Initial laboratory results were significant for leukocytosis with a left shift, lymphopenia, and a microcytic anemia (Table 1). Testing for novel coronavirus 2019 (COVID-19) was negative. Comprehensive metabolic panel (CMP) demonstrated mild hypocalcemia and hypoalbuminemia. C-reactive protein was elevated to 99.4 mg/L (normal: < 8 mg/L) (Table 2). Troponin and B-type natriuretic peptide were normal. Venous blood gas did not demonstrate evidence of acid or base disturbances or carbon dioxide retention. Electrocardiogram (ECG) revealed sinus tachycardia without ST/T wave changes or voltage abnormalities. Chest radiograph (CR) (Fig 2) demonstrated linear atelectasis in the left lower lung field, small bilateral pleural effusions, and a markedly enlarged cardiac silhouette. Given his progressive dyspnea and pleuritic chest pain with associated cardiomegaly on CR, an urgent echocardiogram and pediatric cardiology consult were obtained.
Day of Hospitalization . | |||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|
Clinical Variables . | Reference Range . | 1(ED) . | 1 (PICU) . | 2 . | 3 . | 4 . | 5 . | 6 . | 7 . | 8 . | 9 (at discharge) . |
Complete blood count with differential | |||||||||||
WBC count, K/microliter | 5.00–12.00 | 14.49a | 12.81a | 12.56a | 13.83a | 13.99a | 14.19a | 12.58a | 22.63a | — | 20.19a |
Hemoglobin, g/dL | 11.5–15.5 | 9.3a | 8.9a | 8.6a | 8.8a | 8.1a | 8.3a | 9.7a | 9.4a | — | 10.5a |
Hematocrit, % | 36.0–46.0 | 29.9a | 27.9a | 28.4a | 29.4a | 26.9a | 27.8a | 31.5a | 31.0a | — | 34.8a |
Mean corpuscular volume, fL | 79.0–98.0 | 76.1a | 75.2a | 76.5a | 77.4a | 77.1a | 77.7a | 75.9a | 76.9a | — | 79.3 |
Red cell distribution width, % | 11.5–14.5 | 14.1 | 14.1 | 14.2 | 14.0 | 14.2 | 14.1 | 14.2 | 14.2 | — | 16.0a |
Platelet count, K/microliter | 150–399 | 470a | 404a | 414a | 392 | 457a | 493a | 588a | 626a | — | 512a (normal morphology) |
Neutrophils, K/microliter (%) | 1.55–6.89 (31.0–53.0) | 12.40 (85.7)a | 10.30 (80.3)a | 9.20 (73.3)a | — | 10.62 (75.9)a | 9.61 (67.7)a | — | — | — | 14.0 (69.5)a |
Lymphocytes, K/microliter (%) | 2.00–7.80 (40.0–60.0 | 1.20 (8.3)a | 1.26 (9.8%)a | 1.86 (14.8)a | — | 1.90 (13.6)a | 2.60 (18.3)a | — | — | — | 3.32 (16.4)a |
Monocyte, K/microliter (%) | 0.15–1.30 (3.0–10.0) | 0.73 (5.0) | 1.11 (8.7) | 1.33 (10.6)a | — | 1.30 (9.3) | 1.62 (11.4)a | — | — | — | 1.54 (7.6)a |
Eosinophil, K/microliter (%) | 0.00–0.78 (0.0–6.0) | 0.02 (0.1) | 0.02 (0.2) | 0.08 (0.6) | — | 0.00 (0.0) | 0.01 (0.1) | — | — | — | 0.06 (0.3)a |
Immature granulocytes, K/microliter (%) | 0.00–0.15 (0.0–1.5) | 0.12 (0.8) | 0.10 (0.8) | 0.06 (0.5) | — | 0.16 (1.1) | 0.34 (2.4)a | — | — | — | 1.22 (6.0)a |
Metabolic panel | |||||||||||
Sodium, mmol/L | 137–147 | 136a | 134a | 137 | 140 | 140 | 139 | 138 | — | — | 136a |
Potassium, mmol/L | 3.4–5.3 | 4.4 | 4.1 | 3.7 | 4.1 | 4.2 | 4.0 | 3.6 | — | — | 4.4 |
CO2 total, mmol/L | 20–28 | 23 | 21 | 20 | 18a | 20 | 21 | 20 | — | — | 23 |
Blood urea nitrogen, mg/dL | 5–18 | 8 | 6 | 5 | 11 | 8 | 9 | 11 | — | — | 12 |
Creatinine, mg/dL | 0.52–0.69 | 0.49a | 0.48a | 0.48a | 0.59 | 0.47a | 0.51a | 0.46a | — | — | 0.50a |
Glucose, mg/dL | 60–99 | 95 | 93 | 123a | 151 | 142a | 140a | 109a | — | — | 133a |
Total protein, g/dL | 6.0–8.2 | 8.0 | 7.4 | 7.0 | 7.3 | 6.9 | 7.1 | 6.9 | — | — | 6.4 |
Albumin, g/dL | 3.5–5.0 | 2.9a | 2.5a | 2.2a | 2.1a | 2.0a | 2.3a | 2.4a | — | — | 2.6a |
Bilirubin total, mg/dL | 0.2–1.3 | 0.4 | 0.4 | 0.6 | 0.3 | 0.2 | 0.2 | 0.1a | — | — | 0.2 |
ALT, U/L | 0–40 | 16 | 13 | 9 | 10 | 43a | 41a | 36 | — | — | 25 |
AST, U/L | 3–44 | 20 | 14 | 10 | 19 | 49a | 31a | 19 | — | — | 16 |
Day of Hospitalization . | |||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|
Clinical Variables . | Reference Range . | 1(ED) . | 1 (PICU) . | 2 . | 3 . | 4 . | 5 . | 6 . | 7 . | 8 . | 9 (at discharge) . |
Complete blood count with differential | |||||||||||
WBC count, K/microliter | 5.00–12.00 | 14.49a | 12.81a | 12.56a | 13.83a | 13.99a | 14.19a | 12.58a | 22.63a | — | 20.19a |
Hemoglobin, g/dL | 11.5–15.5 | 9.3a | 8.9a | 8.6a | 8.8a | 8.1a | 8.3a | 9.7a | 9.4a | — | 10.5a |
Hematocrit, % | 36.0–46.0 | 29.9a | 27.9a | 28.4a | 29.4a | 26.9a | 27.8a | 31.5a | 31.0a | — | 34.8a |
Mean corpuscular volume, fL | 79.0–98.0 | 76.1a | 75.2a | 76.5a | 77.4a | 77.1a | 77.7a | 75.9a | 76.9a | — | 79.3 |
Red cell distribution width, % | 11.5–14.5 | 14.1 | 14.1 | 14.2 | 14.0 | 14.2 | 14.1 | 14.2 | 14.2 | — | 16.0a |
Platelet count, K/microliter | 150–399 | 470a | 404a | 414a | 392 | 457a | 493a | 588a | 626a | — | 512a (normal morphology) |
Neutrophils, K/microliter (%) | 1.55–6.89 (31.0–53.0) | 12.40 (85.7)a | 10.30 (80.3)a | 9.20 (73.3)a | — | 10.62 (75.9)a | 9.61 (67.7)a | — | — | — | 14.0 (69.5)a |
Lymphocytes, K/microliter (%) | 2.00–7.80 (40.0–60.0 | 1.20 (8.3)a | 1.26 (9.8%)a | 1.86 (14.8)a | — | 1.90 (13.6)a | 2.60 (18.3)a | — | — | — | 3.32 (16.4)a |
Monocyte, K/microliter (%) | 0.15–1.30 (3.0–10.0) | 0.73 (5.0) | 1.11 (8.7) | 1.33 (10.6)a | — | 1.30 (9.3) | 1.62 (11.4)a | — | — | — | 1.54 (7.6)a |
Eosinophil, K/microliter (%) | 0.00–0.78 (0.0–6.0) | 0.02 (0.1) | 0.02 (0.2) | 0.08 (0.6) | — | 0.00 (0.0) | 0.01 (0.1) | — | — | — | 0.06 (0.3)a |
Immature granulocytes, K/microliter (%) | 0.00–0.15 (0.0–1.5) | 0.12 (0.8) | 0.10 (0.8) | 0.06 (0.5) | — | 0.16 (1.1) | 0.34 (2.4)a | — | — | — | 1.22 (6.0)a |
Metabolic panel | |||||||||||
Sodium, mmol/L | 137–147 | 136a | 134a | 137 | 140 | 140 | 139 | 138 | — | — | 136a |
Potassium, mmol/L | 3.4–5.3 | 4.4 | 4.1 | 3.7 | 4.1 | 4.2 | 4.0 | 3.6 | — | — | 4.4 |
CO2 total, mmol/L | 20–28 | 23 | 21 | 20 | 18a | 20 | 21 | 20 | — | — | 23 |
Blood urea nitrogen, mg/dL | 5–18 | 8 | 6 | 5 | 11 | 8 | 9 | 11 | — | — | 12 |
Creatinine, mg/dL | 0.52–0.69 | 0.49a | 0.48a | 0.48a | 0.59 | 0.47a | 0.51a | 0.46a | — | — | 0.50a |
Glucose, mg/dL | 60–99 | 95 | 93 | 123a | 151 | 142a | 140a | 109a | — | — | 133a |
Total protein, g/dL | 6.0–8.2 | 8.0 | 7.4 | 7.0 | 7.3 | 6.9 | 7.1 | 6.9 | — | — | 6.4 |
Albumin, g/dL | 3.5–5.0 | 2.9a | 2.5a | 2.2a | 2.1a | 2.0a | 2.3a | 2.4a | — | — | 2.6a |
Bilirubin total, mg/dL | 0.2–1.3 | 0.4 | 0.4 | 0.6 | 0.3 | 0.2 | 0.2 | 0.1a | — | — | 0.2 |
ALT, U/L | 0–40 | 16 | 13 | 9 | 10 | 43a | 41a | 36 | — | — | 25 |
AST, U/L | 3–44 | 20 | 14 | 10 | 19 | 49a | 31a | 19 | — | — | 16 |
ALT, alanine aminotransferase; AST, aspartate aminotransferase; CBC, complete blood count; PICU, pediatric intensive care unit; WBC, white blood cell; —, not applicable.
Abnormal values.
Day of Hospitalization . | ||||||||
---|---|---|---|---|---|---|---|---|
Inflammatory markers | ||||||||
Clinical variables | Reference Range | 1 (ED) | 1 (PICU) | 2 | 3 | 4 | 5 | 6 |
C-reactive protein (mg/L) | 0.0–8.0 | 99.4a | — | 114.3a | — | 53.5a | 14.7a | — |
Lactate dehydrogenase (U/L) | 110–240 | — | 203 | — | — | — | — | — |
Uric acid (mg/dL) | 2.0–5.5 | — | — | 4.5 | — | — | — | — |
Erythrocyte sedimentation rate (mm/hr) | 0–17 | — | — | 86a | 106a | 105a | — | 60a |
Aldolase (U/L) | 3.3–10.3 | — | — | 21.1a | — | — | — | — |
Creatine kinase (U/L) | 10–205 | — | — | 122.0 | — | — | — | — |
Cardiac markers | ||||||||
Troponin I (ng/mL) | 0.00–0.09 | < 0.01 | < 0.01 | < 0.01 | < 0.01 | < 0.01 | — | < 0.01 |
Brain natriuretic peptide (pg/mL) | 0–100 | 25 | — | 85 | 119a | 123a | — | 280a |
Day of Hospitalization . | ||||||||
---|---|---|---|---|---|---|---|---|
Inflammatory markers | ||||||||
Clinical variables | Reference Range | 1 (ED) | 1 (PICU) | 2 | 3 | 4 | 5 | 6 |
C-reactive protein (mg/L) | 0.0–8.0 | 99.4a | — | 114.3a | — | 53.5a | 14.7a | — |
Lactate dehydrogenase (U/L) | 110–240 | — | 203 | — | — | — | — | — |
Uric acid (mg/dL) | 2.0–5.5 | — | — | 4.5 | — | — | — | — |
Erythrocyte sedimentation rate (mm/hr) | 0–17 | — | — | 86a | 106a | 105a | — | 60a |
Aldolase (U/L) | 3.3–10.3 | — | — | 21.1a | — | — | — | — |
Creatine kinase (U/L) | 10–205 | — | — | 122.0 | — | — | — | — |
Cardiac markers | ||||||||
Troponin I (ng/mL) | 0.00–0.09 | < 0.01 | < 0.01 | < 0.01 | < 0.01 | < 0.01 | — | < 0.01 |
Brain natriuretic peptide (pg/mL) | 0–100 | 25 | — | 85 | 119a | 123a | — | 280a |
Abnormal values; —, not applicable.
Question 1: Dr Murphy, (pediatric cardiologist) as the consultant in the ED, what are your initial thoughts on the diagnosis and management of this patient?
The incidence of cardiac disease in children presenting to emergency departments with the chief complaint of chest pain is <1%.1 Although cardiac disease is rare in previously healthy children, early suspicion for, and recognition of signs and symptoms associated with acute myocardial infarct, arrhythmias, and myo-pericardial disease, are essential because these disease entities can rapidly become life threatening. The acute diagnoses that are highest on my differential are myocarditis and pericarditis. Myocarditis may present nonspecifically with symptoms such as flu-like illness or chest pain and can progress quickly to heart failure and cardiogenic shock. Similarly to our patient, children often present with shortness of breath, dyspnea on exertion, and decreased appetite. A gallop rhythm may be noted on physical exam. Elevated cardiac biomarkers can confirm the diagnosis of cardiac injury; however, normal levels like those in our patient, do not exclude the diagnosis. ECG can be normal, but common ECG findings include sinus tachycardia, ST-wave and T-wave abnormalities, low voltage readings, and less frequently atrioventricular or bundle branch block.2 Pericarditis commonly presents with chest pain that is relieved by leaning forward and exacerbated by coughing, lying down, or with inspiration. The cardiac examination may be normal or may demonstrate a precordial friction rub with “muffled” or distant heart sounds on auscultation. An ECG will generally demonstrate widespread ST-segment elevation and PR depression but can also be normal.2 One may also see electrical alternans which is variation in the amplitude of the QRS complexes. This is thought to be caused by electrical signals detected over varying distances in the pericardial sac as the heart swings in an anterior-posterior motion though the fluid.3 A transthoracic echocardiogram is the next best step to distinguish pericarditis from myocarditis, as well as evaluate for other causes of cardiomegaly such as valvular disease, cardiomyopathy, pulmonary disease leading to right-sided cardiac enlargement, and malignancy of the mediastinum. An urgent assessment of cardiac function will also guide the initial resuscitation while we begin an investigation into the underlying etiology.
Dr Hall:
A transthoracic echocardiogram (Fig 3) revealed a large pericardial effusion measuring ∼3.5 cm with right atrial and right ventricular chamber collapse. Pulsed-wave Doppler of the mitral valve showed a >25% difference between inspiration and expiration, an echocardiographic representation of tamponade physiology. Ejection fraction was estimated at 66%. The patient was admitted to the pediatric ICU for further management.
Question 2: Dr Wilkerson (pediatric intensivist), how does one identify cardiac tamponade physiology and what would be the next steps to stabilize this patient?
Cardiac tamponade is a medical emergency that occurs when the amount of fluid in the pericardial sac reaches a level that compromises cardiac function. The compression of the heart by the pericardial fluid reduces ventricular filling during diastole and obstructs both venous return and cardiac output. Early signs of cardiac tamponade are tachypnea, tachycardia, narrow pulse pressure, and elevated systemic venous pressure. Examination findings may include “muffled” heart sounds or a friction rub on cardiac auscultation which are caused by the presence of excess fluid surrounding the heart. Neck vein distention secondary to elevated venous pressure, and anxiety or light-headedness from inadequate oxygen delivery to the brain may be present. A clinical hallmark of cardiac tamponade is pulsus paradoxus. During the inspiratory phase of respiration, there is normally a decrease in systolic arterial pressure. With cardiac tamponade, this physiologic phenomenon is exaggerated because of reduced left ventricular stroke volume during inspiration. Pulsus paradoxus is defined as a >10 mm Hg decrease in systolic arterial pressure during inspiration.4 Hypotension is a late finding synonymous with cardiogenic shock.
The next step in management of this patient is expansion of the intravascular volume as hypovolemia reduces cardiac output, thereby exacerbating the pathophysiology of this condition. Urgent fluid administration is only a temporizing measure while access is obtained for the use of inotropes to support cardiac output and arrangements are made for the definitive treatment of cardiac tamponade, emergent pericardiocentesis with or without placement of a percutaneous pericardial catheter.
Dr Hall:
Echocardiography-guided pericardiocentesis was performed extracting 300 mL of serosanguinous pericardial fluid, and a pigtail drain was placed. Repeat echocardiogram demonstrated a reduced volume of pericardial fluid and good biventricular function. Pericardial fluid cytology was grossly bloody with an 83% neutrophilic predominance (Table 3). Gram stain was negative. Although we had low suspicion for bacterial pericarditis given the lack of fevers and negative gram stain, ceftriaxone was started for empirical antibacterial coverage because of the high incidence of morbidity and mortality associated with bacterial pericarditis. In addition, the patient was started on methylprednisolone 1 mg/kg/dose twice daily for treatment of pericarditis.
Pericardial Fluid . | ||
---|---|---|
Clinical Variable . | Reference Range . | Value . |
Total protein (g/dL) | — | 5.9 |
WBC count (/microliter) | — | 7994 |
RBC count (/microliter) | — | 119 000 |
Fluid color | None | Slightly Xanthochromica |
Fluid clarity | Clear | Cloudya |
Fluid gross blood | None | Markeda |
Mononuclear absolute (/microliter) (%) | — | 1343 (16.8) |
Polymorphonuclear absolute (/microliter) (%) | — | 6651 (83.2) |
Pericardial Fluid . | ||
---|---|---|
Clinical Variable . | Reference Range . | Value . |
Total protein (g/dL) | — | 5.9 |
WBC count (/microliter) | — | 7994 |
RBC count (/microliter) | — | 119 000 |
Fluid color | None | Slightly Xanthochromica |
Fluid clarity | Clear | Cloudya |
Fluid gross blood | None | Markeda |
Mononuclear absolute (/microliter) (%) | — | 1343 (16.8) |
Polymorphonuclear absolute (/microliter) (%) | — | 6651 (83.2) |
RBC, red blood cell; —, not applicable.
Abnormal values.
Question 3: Dr Murphy, what is the differential diagnosis for acute pericarditis?
According to the literature, between 40% to 86% of cases of pediatric pericarditis are considered idiopathic.5 The causes of pericarditis in children can be divided into infectious and noninfectious causes, with the latter including immunoreactive or autoimmune, metabolic, neoplastic, and traumatic causes. When complicated, as in the case of large effusions or those presenting with tamponade physiology, an extensive workup is often warranted.
Dr Hall:
The patient’s blood gas did not demonstrate any acid or base disturbances and the CMP was only significant for mild hypocalcemia and hypoalbuminemia (Table 1). Thyroid stimulating hormone and free thyroxine were also normal, allowing us to rule out metabolic causes such as uremia or thyroid pathology. Although the pericardial fluid was grossly bloody, our patient denied any recent trauma to the chest. It is also important to consider inadvertent injury to the cardiac muscle or perforation during the pericardiocentesis.
Question 4: Dr Wohrley (pediatric infectious diseases), what infectious causes would you be most concerned for in a patient with this pericardial fluid analysis?
The presence of a neutrophilic predominance within the fluid is of unclear significance, but given lack of fever on presentation and negative gram stain, typical bacterial causes are less likely. The presence of significant hemorrhagic pericardial effusion in the setting of a 20 pound weight loss and cough illness earlier in the year make tuberculosis (TB) disease a possibility.
Infectious pericarditis can be classified as benign, purulent, or granulomatous. Benign pericarditis is most often a consequence of viral infections or in the setting of postpericardiotomy syndromes. Of the viral causes of pericarditis, enteroviruses predominate, particularly coxsackieviruses. Other less commonly implicated viral causes include adenovirus, parvovirus B19, HIV, Epstein-Barr virus (EBV), cytomegalovirus (CMV), other herpesviruses, and COVID-19. Even less common viruses causing pericarditis include influenza A or B, hepatitis B or C, mumps, and lymphocytic choriomeningitis viruses.6 CMV as a cause of pericarditis occurs in the setting of immunocompromised hosts and among those infected with HIV.6 Although viral causes of pericarditis predominate in children, a lack of viral prodrome and other associated viral symptoms in the current illness argue against viral causes in this case. Purulent pericarditis is most commonly caused by Staphylococcus aureus, group A Streptococcus, the Streptococcus milleri group, Salmonella species, Pseudomonas aeruginosa, and other gram negative pathogens. Haemophilus influenzae type b, Streptococcus pneumoniae, and Neisseria meningitidis are less prevalent in the postvaccine era.6 Granulomatous pericarditis is because of mycobacterial or fungal etiologies. TB is the most common etiology of pericarditis worldwide, largely in developing countries where TB is endemic.6 When considering TB, obtaining a detailed history including travel, origin of birth, and exposure to TB-infected or high-risk individuals is important. Finally, although much less likely, eliciting a travel history for exposure to endemic fungi or activities increasing this risk is important. This patient’s exposure history was significant only for exposure to geese, rabbits, and possums in his yard, without direct contact. There was no personal or family history of travel and no travelers visited from TB endemic areas. Should the patient have had a personal history of COVID 19 or had close exposures, multisystem inflammatory syndrome in children would have been a consideration. Although our patient had not been vaccinated against COVID-19 virus, it is also important to consider post vaccination myocarditis as a possible etiology in adolescent males who have been vaccinated.7
Dr Hall:
An acid fast bacilli smear and TB polymerase chain reaction (PCR) of the pericardial fluid as well as a quantiFERON-TB Gold test and purified protein derivative were obtained, and our patient was placed in airborne isolation while awaiting results. Enterovirus and adenovirus PCRs of blood and pericardial fluid, as well as a respiratory pathogen panel, HIV antigen or antibody, EBV viral capsid antigen (serologies and blood PCR), CMV serologies, parvovirus B19 (serologies and blood PCR), urine histoplasma antigen, Cryptococcus antigen in pericardial fluid, and Mycoplasma serologies were obtained.
In addition to the pericardial effusion, the patient’s symptoms of fatigue and weight loss over several months prompted an investigation into possible oncologic etiologies. His mother initially attributed weight loss to improvement in diet but also stated that he had decreased appetite more recently. Although he denied a history of easy bleeding, bruising, or bone pain, we considered the possibility of a malignant pericardial effusion.
Question 5: Dr Kent (pediatric hematology or oncology), what are the oncologic considerations for a large pericardial effusion in this clinical presentation?
A 2008 study done by Kühn et al on 116 pediatric patients with pericardial effusions at Boston Children's Hospital found that 39% had an underlying neoplastic disease, 37% had unclear etiology and were classified as idiopathic, 9% had collagen vascular disease, 8% had renal disease, 3% had bacterial infections, and 2% had HIV.8 Another retrospective study done on 173 patients (including adults) presenting with pericardial effusion requiring pericardiocentesis found that 7.5% had an unrecognized neoplastic disease, most commonly lung cancer, but which also included acute leukemia, Hodgkin disease, and non-Hodgkin lymphoma.9
This patient’s complete blood count demonstrated mild leukocytosis with lymphopenia, microcytic normochromic anemia, and an elevated platelet count, suggestive of anemia of acute inflammation but not acute leukemia. He had no hepatosplenomegaly or lymphadenopathy on exam or CR. He had a normal lactate dehydrogenase making hematologic malignancy unlikely (Table 2). However, given his history of premature pubarche we recommended obtaining markers for germ cell tumors, which are known to present with pericardial effusions, including: carcinoembryonic antigen, α-fetoprotein (AFP), β-human chorionic gonadotropin, carbohydrate antigen 19-9, carbohydrate antigen 125, luteinizing hormone (LH), follicle stimulating hormone, and testosterone. Additionally, we recommended sending the pericardial fluid for hematopathology review with immunohistochemistry, flow cytometry to look for an abnormal clone, and cytogenetics to look for known genetic translocations found commonly in childhood leukemia or lymphoma [t(2:5), t(9;22), 11q23, t(1;19)].
Lastly nonmalignant but devastating conditions such as lymphoproliferative hemophagocytosis, macrophage activation syndrome, histiocytosis, hemolytic uremic syndrome, thrombotic thrombocytopenic purpura, disseminated intravascular coagulation, sinus histiocytosis (Rosai Dorfman disease), and severe hemolytic anemia can present in this way. We recommended sending a coagulation profile, iron studies, and Lupus anticoagulant.
Dr Hall:
Activated partial thromboplastin clotting time, prothrombin time, prothrombin time and international normalized ratio, and thrombin all returned normal. D dimer was elevated to 9.63 mg/L FEU (reference range: 0.00–0.60). Fibrinogen was elevated to 646 mg/dL (reference range 190–295). These results, along with iron studies (low iron, normal ferritin, low total iron binding capacity, low transferrin, and low transferrin saturation), all suggested acute inflammation but did not support any of the above specific diagnoses.
The patient’s pericardial drain continued to have serosanguinous output, accumulating an additional 774mL over 48 hours since the initial aspiration. He developed a fever on hospital day 2 (Fig 1). At that time a blood culture, urinalysis, and urine culture were obtained. Urinalysis revealed microscopic hematuria and proteinuria but no evidence of infection (Table 4).
Day of Hospitalization . | |||||
---|---|---|---|---|---|
Urinalysis . | |||||
Clinical Variables . | Reference Range . | 1 (PICU) . | 2 . | 3 . | |
Specific gravity | 1.002 –1.030 | 1.017 | 1.023 | 1.033 | |
Protein (mg/dL) | < 20 | 100a | 100a | 200a | |
Glucose (mg/dL) | < 30 | < 30 | < 30 | < 30 | |
Blood | Negative | 1+a | 2+a | 2+a | |
Nitrite | Negative | negative | negative | negative | |
Urobilinogen (mg/dL) | < 2 | < 2 | 4a | 3a | |
Leukocyte esterase (Leu/microliter) | Negative | negative | negative | negative | |
RBC (/hpf) | 0–5 | 14a | 29a | 58a | |
WBC (/hpf) | 0–5 | 8a | 17a | 17a | |
Bacteria | None | Rare | Few | Few | |
Squamous epithelial cells (/lpf) | 0–5 | 0 | 5 | 3 | |
Hyaline casts (/Ipf) | 0–5 | — | 8a | 10a |
Day of Hospitalization . | |||||
---|---|---|---|---|---|
Urinalysis . | |||||
Clinical Variables . | Reference Range . | 1 (PICU) . | 2 . | 3 . | |
Specific gravity | 1.002 –1.030 | 1.017 | 1.023 | 1.033 | |
Protein (mg/dL) | < 20 | 100a | 100a | 200a | |
Glucose (mg/dL) | < 30 | < 30 | < 30 | < 30 | |
Blood | Negative | 1+a | 2+a | 2+a | |
Nitrite | Negative | negative | negative | negative | |
Urobilinogen (mg/dL) | < 2 | < 2 | 4a | 3a | |
Leukocyte esterase (Leu/microliter) | Negative | negative | negative | negative | |
RBC (/hpf) | 0–5 | 14a | 29a | 58a | |
WBC (/hpf) | 0–5 | 8a | 17a | 17a | |
Bacteria | None | Rare | Few | Few | |
Squamous epithelial cells (/lpf) | 0–5 | 0 | 5 | 3 | |
Hyaline casts (/Ipf) | 0–5 | — | 8a | 10a |
Abnormal values.
On hospital day 2, he became hypertensive with blood pressures ranging from 140 to 160 per 70 to 90 (98th percentile for age). This was initially attributed to the receipt of steroids and his intravascular fluid vol. His physical exam was significant at that time for mild periorbital edema, but no pretibial edema or jugular venous distension. We treated the hypertension with hydralazine, followed by 1 dose of intravenous furosemide with minimal improvement. Repeat urinalysis was significant for subnephrotic range proteinuria, microscopic hematuria, and pyuria. Urine protein:creatinine ratio was 1.2 mg/dL concerning for significant proteinuria. Given the need to balance blood pressure management with maintaining adequate preload in the setting of a persistent pericardial effusion, in addition to possible primary renal pathology, he decision was made to consult pediatric nephrology for further guidance.
Question 6: Dr Jandeska (pediatric nephrologist), what are your recommendations based on our patient’s renal findings?
The patient’s clinical picture is concerning for glomerulonephritis with a normal glomerular filtration rate. SBP greater than 130mmHg can be treated with labetalol 10mg every 4 hours as needed. On further history, the patient’s mother shared that there was a family history of systemic lupus erythematosus (SLE) in his maternal grandmother. Although the patient initially denied any of the more common symptoms suggestive of autoimmune disease such as rash, fever, lymphadenopathy, photosensitivity, conjunctivitis, mouth ulcers, neurologic changes, psychiatric concerns, eye complaints, or arthralgias, his mother did note that about a month before she noticed her son was generally fatigued and would complain of vague leg pain without swelling. This history in conjunction with weight loss, anemia, concern for nephrotic syndrome with hypoalbuminemia on CMP, new onset fever, and family history of autoimmune disease, prompted investigation of SLE as a possible cause of his pericardial effusion and other symptoms.
Dr Hall:
The patient’s antinuclear antibodies (ANA) was positive, and double stranded DNA (dsDNA), anti-Smith antibody, and anti-ribonucleoprotein (anti-RNP) titers all returned elevated. Although serum complement component 3 (C3) was normal, complement component 4 (C4) was depressed (Table 5). In the absence of pediatric rheumatology at our institution, we consulted allergy and immunology after these laboratories resulted.
Immunologic Work-Up . | ||
---|---|---|
Clinical Variables . | Reference Range . | Value . |
Rheumatoid factor (IU/mL) | 0–29 | < 15 |
CCP antibodies (units) | 0–19 | 6 |
Anti-Ro or SSA antibodies | 0–0.9 | 0.9 |
Anti-La or SSB antibodies | 0–0.9 | < 0.2 |
ANA | Negative | Positivea |
Anti-RNP and Anti SM antibodies | 0–0.9 | > 8.0a |
Anti-SM antibodies | 0–0.9 | > 8.0a |
Antidouble stranded DNA (IU/mL) | 0–4.0 | 183.0a |
Immunofluorescent ANA | < 1:40 | ≥ 1:1280a |
C3 complement (mg/dL) | 88–203 | 118 |
C4 complement (mg/dL) | 13.0–49.0 | 10.8a |
Anti-Jo1 antibody | 0.0–0.9 | < 0.2 |
Centromere autoantibodies | 0.0–0.9 | < 0.2 |
Myeloperoxidase antibody (U/mL) | 0.0–0.9 | < 9.0 |
Proteinase 3 antibody (U/mL) | 0.0–3.5 | < 3.5 |
C-ANCA titer | Negative: < 1:20 | < 1:20 |
P-ANCA titer | Negative: < 1:20 | > 1:640a |
Atypical P-ANCA titer | Negative: < 1:20 | < 1:20 |
Immunologic Work-Up . | ||
---|---|---|
Clinical Variables . | Reference Range . | Value . |
Rheumatoid factor (IU/mL) | 0–29 | < 15 |
CCP antibodies (units) | 0–19 | 6 |
Anti-Ro or SSA antibodies | 0–0.9 | 0.9 |
Anti-La or SSB antibodies | 0–0.9 | < 0.2 |
ANA | Negative | Positivea |
Anti-RNP and Anti SM antibodies | 0–0.9 | > 8.0a |
Anti-SM antibodies | 0–0.9 | > 8.0a |
Antidouble stranded DNA (IU/mL) | 0–4.0 | 183.0a |
Immunofluorescent ANA | < 1:40 | ≥ 1:1280a |
C3 complement (mg/dL) | 88–203 | 118 |
C4 complement (mg/dL) | 13.0–49.0 | 10.8a |
Anti-Jo1 antibody | 0.0–0.9 | < 0.2 |
Centromere autoantibodies | 0.0–0.9 | < 0.2 |
Myeloperoxidase antibody (U/mL) | 0.0–0.9 | < 9.0 |
Proteinase 3 antibody (U/mL) | 0.0–3.5 | < 3.5 |
C-ANCA titer | Negative: < 1:20 | < 1:20 |
P-ANCA titer | Negative: < 1:20 | > 1:640a |
Atypical P-ANCA titer | Negative: < 1:20 | < 1:20 |
Anti-RNP, anti-ribonucleoprotein; ANCA, antineutrophil cytoplasmic antibodies; SSA, Sjogren syndrome A; SSB, Sjogren syndrome B.
Abnormal values.
Question 7: Dr Bandi (Allergy or Immunology), with the above laboratory results, what additional information should we obtain to confirm our suspicion of SLE?
Although pediatric SLE does not have its own diagnostic criteria, the most widely used and accepted criteria have been the 1997 American College of Rheumatology (ACR) criteria. Additional criteria have been published since, including the Systemic Lupus International Collaborating Clinics (SLICC) criteria in 2012 and the European League Against Rheumatism (EULAR) or ACR criteria in 2019 that included entry criteria (an elevated ANA).10 In 1 study, the 2019 EULAR or ACR criteria and the 2012 SLICC criteria were found to be more sensitive than the 1997 ACR criteria. There were, however, no significant differences in sensitivity and specificity between the 2012 SLICC and the 2019 EULAR of ACR criteria.11 Even without further laboratory review, the patient meets ACR criteria for the diagnosis of SLE given he meets 6 out of 11 criteria (4 out of 11 required for diagnosis): serositis in the form of pericarditis, lymphopenia, proteinuria, positive ANA, and positive dsDNA, as well as anti-Sm.12
A normal C3 and low C4 can be seen in SLE, which is the case in our patient. You can also see this in mixed cryoglobulinemia (sometimes associated with hepatitis) and membranous glomerulonephritis. Also on the allergy or immunology spectrum, both hereditary and acquired angioedema will have a low C4 in the setting of normal C3.
Treatment of a presumed diagnosis of SLE was initiated first by increasing methylprednisolone dosing to 1 g daily and a thorough autoimmune workup was completed (Table 5).
Final Diagnosis: Dr Hall:
Lactate dehydrogenase and tumor markers carcinoembryonic antigen, AFP, beta human chorionic gonadotropin, and carbohydrate antigen 19-9 all returned normal, reducing our suspicion for a germ cell tumor. Carbohydrate antigen 125 was elevated to 65 U/mL (reference range 0–31 U/mL); however, this can also be seen with large pericardial effusions, regardless of etiology.13 Flow cytometry did not show a clonal population, nor stain for acute lymphoblastic leukemia or acute myelogenous leukemia. All initial and subsequent infectious studies were negative, with the exception of EBV PCR, EBV IgG, and CMV IgG. The pericardial fluid culture remained with no growth throughout the hospital course (Table 6). In light of his rheumatologic evaluation (including positive ANA, dsDNA, Anti-Smith, and anti-ribonucleoprotein), an autoimmune process such as SLE was considered the most likely diagnosis.
Infectious work-up . | ||
---|---|---|
Clinical Variables . | Reference Range . | Value . |
COVID-19 | — | Not detected |
Respiratory pathogen panel PCR, nasopharynx | — | None detected |
HIV antigen or antibody | — | Not detected |
Serum EBV DNA PCR (copies/mL) | < 300 | 487a |
EBV viral capsid antigen IgG | 0–0.8 | > 8.0a |
EBV viral capsid antigen IgM | 0–0.8 | < 0.2 |
Serum CMV DNA quantitation (copies/mL) | < 300 | <300 |
CMV IgG | 0–0.8 | 3.8a |
CMV IgM | 0–0.8 | 0.2 |
HSV 1 and 2 IgG | 0–0.8 | 0.2, < 0.2 |
Cryptococcal antigen | — | None detected |
HHV 6 PCR (copies/mL) | < 500 | < 500 |
Parvovirus B19 DNA, PCR | — | Negative |
Parvovirus B19 IgG | 0.0–0.8 | 2.9a |
Parvovirus B19 IgM | 0.0–0.8 | 0.1 |
M pneuomniae IgG (U/mL) | 0 –99 | < 100 |
M pneuomniae IgM (U/mL) | 0–769 | < 770 |
Serum lysozyme (ug/mL) | 3.0–12.8 | 5.3 |
Serum adenosine deaminase (microns/L) | 0.00–15.00 | 8.10 |
Quantiferon TB Gold | — | Negative |
Blood cultures | — | No growth after 5 d of incubation |
Urine culture | — | No growth |
Urine histoplasma antigen | — | None detected |
Pericardial fluid HSV 1 and 2 PCR | — | Negative |
Pericardial fluid adenovirus PCR | Negative | |
Pericardial fluid enterovirus PCR | — | Negative |
Pericardial fluid MTB PCR | — | Not detected |
Pericardial fluid gram stain and fluid aerobic culture | — | Few WBCs, no organisms seen, no growth on culture |
Pericardial fluid anaerobic culture | — | No anaerobes isolated after 5 d of inoculation |
Pericardial fluid AFB culture with smear, fluorochrome stain | — | No acid-fast bacilli isolated after 8 wk, no acid fast bacilli seen with stain |
Pericardial fluid fungal culture, smear, and calcofluor stain | — | No fungi isolated after 4 wk, no fungal elements seen with stain |
Infectious work-up . | ||
---|---|---|
Clinical Variables . | Reference Range . | Value . |
COVID-19 | — | Not detected |
Respiratory pathogen panel PCR, nasopharynx | — | None detected |
HIV antigen or antibody | — | Not detected |
Serum EBV DNA PCR (copies/mL) | < 300 | 487a |
EBV viral capsid antigen IgG | 0–0.8 | > 8.0a |
EBV viral capsid antigen IgM | 0–0.8 | < 0.2 |
Serum CMV DNA quantitation (copies/mL) | < 300 | <300 |
CMV IgG | 0–0.8 | 3.8a |
CMV IgM | 0–0.8 | 0.2 |
HSV 1 and 2 IgG | 0–0.8 | 0.2, < 0.2 |
Cryptococcal antigen | — | None detected |
HHV 6 PCR (copies/mL) | < 500 | < 500 |
Parvovirus B19 DNA, PCR | — | Negative |
Parvovirus B19 IgG | 0.0–0.8 | 2.9a |
Parvovirus B19 IgM | 0.0–0.8 | 0.1 |
M pneuomniae IgG (U/mL) | 0 –99 | < 100 |
M pneuomniae IgM (U/mL) | 0–769 | < 770 |
Serum lysozyme (ug/mL) | 3.0–12.8 | 5.3 |
Serum adenosine deaminase (microns/L) | 0.00–15.00 | 8.10 |
Quantiferon TB Gold | — | Negative |
Blood cultures | — | No growth after 5 d of incubation |
Urine culture | — | No growth |
Urine histoplasma antigen | — | None detected |
Pericardial fluid HSV 1 and 2 PCR | — | Negative |
Pericardial fluid adenovirus PCR | Negative | |
Pericardial fluid enterovirus PCR | — | Negative |
Pericardial fluid MTB PCR | — | Not detected |
Pericardial fluid gram stain and fluid aerobic culture | — | Few WBCs, no organisms seen, no growth on culture |
Pericardial fluid anaerobic culture | — | No anaerobes isolated after 5 d of inoculation |
Pericardial fluid AFB culture with smear, fluorochrome stain | — | No acid-fast bacilli isolated after 8 wk, no acid fast bacilli seen with stain |
Pericardial fluid fungal culture, smear, and calcofluor stain | — | No fungi isolated after 4 wk, no fungal elements seen with stain |
AFB, acid fast bacilli; HHV, human herpes virus; HSV, herpes simplex virus; MT, mycobacterium tuberculosis; —, not applicable.
Abnormal values.
The pericardial drain was removed on hospital day 4 after output became minimal and repeat echocardiogram demonstrated trace pericardial effusion.
A diagnostic or prognostic renal biopsy revealed proliferative glomerulonephritis in 4 out of 40 glomeruli, and 2 cellular crescents were identified. There were also areas of thickened glomerular basement membrane with subepithelial complexes observed, consistent with a certain degree of membranous nephropathy. Tubulointerstitial infiltrates were illustrated in 15% of the cortex. No chronicity was observed. Overall, these findings were consistent with focal proliferative lupus nephritis with glomerular necrosis and membranous lupus nephritis, International Society of Nephrology and the Renal Pathology Society (ISN/RPS) Class III (A) + Class V (Fig 4).
Treatment of our patient’s proliferative glomerulonephritis was initiated per the National Institutes of Health protocol with cyclophosphamide (650 mg/m2 monthly for 6 doses) and prednisone 30 mg twice daily (to be weaned at 4–6 weeks). He was discharged on trimethoprim or sulfamethoxazole every Monday, Wednesday, and Friday for Pneumocystis jirovecii pneumonia prophylaxis, and his hypertension was managed with valsartan 40 mg daily and nifedipine extended release 30 mg twice daily.
Discussion
In the setting of such a broad differential diagnosis, the assessment of a pericardial effusion large enough to cause tamponade physiology in a pediatric patient merits a multidisciplinary approach. Initially, an autoimmune etiology was not high on the differential given our patient’s age and sex. However, up to 20% of individuals with SLE will have onset of their disease before age 18.14 Additionally, the female to male ratio in pediatric SLE differs from the adult ratio of 9:1, with a 4:3 ratio reported with disease onset during the first decade of life, and 4:1 during the second decade.15 On learning of an SLE diagnosis in our patient’s maternal grandmother, along with onset of fever, hypertension, and proteinuria, we began to investigate autoimmune etiologies. It is suggested that a family history is present in as many as 47% of all SLE cases and a higher genetic risk has been suggested in pediatric SLE as compared to adult SLE.16,17
Although the most common presentation of SLE in children tends to be more nonspecific symptoms such as fatigue, fever, lymphadenopathy, and weight loss, pericarditis is present in about 17 to 33% of SLE patients (regardless of age) and is the most commonly diagnosed cardiac manifestation of SLE.18 It is important to note that our patient’s presentation remains rare. In a recent comprehensive analysis of pediatric cardiac tamponade cases, Maharaj et al noted only 1 male patient out of a total of 13 cases of pediatric SLE initially presenting with cardiac tamponade.19
Our patient’s weight loss and hemorrhagic pericardial fluid led us to conduct a full TB evaluation before suspecting an autoimmune etiology. In a review by Arabi et al, 8 out of 11 children with cardiac tamponade as a presenting symptom of SLE had sanguinous or frankly bloody fluid, and 3 of those patients were empirically treated for TB until ultimately having negative test results.20 This highlights the importance of having SLE on the differential in a patient with grossly bloody pericardial fluid. All 11 of the reported cases in this study had elevated anti-dsDNA titer, ANA titer, or both. In a study of 418 patients with juvenile-onset SLE, Massias et al demonstrated that prepubertal children demonstrate lower rates of ANA positivity and lower median anti-dsDNA titers, but by adolescence, only 3% of patients are ANA negative.14 This supports screening with ANA in a pubertal child who presents with cardiac tamponade. Low serum complement C4 level is found to be predictive of progression to cardiac tamponade in a series of adolescent SLE patients, and our patient did have a low C4.19,21
Although a relatively rare pediatric diagnosis and presentation, this case highlights the importance of considering SLE when a child presents with a large pericardial effusion. At a minimum, the medical team should obtain an appropriate autoimmune history and review of systems and should strongly consider basic autoimmune screening laboratories, including ANA, dsDNA, and complement levels. It is well established that children and adolescents diagnosed with SLE generally have a more severe disease presentation, develop disease damage more quickly than adults, and have a higher overall burden of disease over their lifetimes.22 These recommendations could aid in early detection, earlier initiation of treatment and, therefore, better outcomes for pediatric patients with SLE.
Acknowledgments
Thank you to Dr David Cimbaluk, pathologist on case, who provided histologic slides and Dr Joshua Murphy, cardiologist on case, for echocardiogram images. Thank you to Dr Colleen Nash who contributed to the editing of this publication.
Dr Hall led the initial writing of the manuscript, recruited various specialists for writing the manuscript, revised the manuscript, and was involved in the care of the patient; Drs Rosado, Wohrley, Kent, Bandi, Murphy, and Jandeska contributed to the writing of the manuscript and were involved in the care of the patient; Dr Wilkerson led the writing of the final manuscript, revised the manuscript, and was involved in the care of the patient; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
FUNDING: No external funding.
CONFLICT OF INTEREST DISCLOSURES: The authors have no conflicts of interest relevant to this article to disclose.
- ACR
American College of Rheumatology
- AFB
acid fast bacilli
- AFP
α-fetoprotein
- ANA
antinuclear antibodies
- CR
chest radiograph
- C3
complement component 3
- C4
complement component 4
- CMP
comprehensive metabolic panel
- CMV
cytomegalovirus
- COVID-19
coronavirus disease 2019
- dsDNA
double stranded DNA
- EBV
Epstein-Barr virus
- ED
emergency department
- ECG
electrocardiogram
- EULAR
European League Against Rheumatism
- LH
luteinizing hormone
- TB
tuberculosis
- SLE
systemic lupus erythematosus
- SLICC
Systemic Lupus International Collaborating Clinics
Comments