Cardiac arrest in a healthy child due to paradoxical embolus across a previously unrecognised sinus venosus defect

Abstract

A previously healthy, preadolescent female suffered an unwitnessed cardiac arrest with prompt return of circulation following bystander initiated resuscitation. Workup demonstrated the cause of her cardiac arrest to be distal left anterior descending coronary artery occlusion with small apical left ventricular transmural myocardial infarction, from a paradoxical embolus traversing a previously undiagnosed large sinus venous defect. This case demonstrates the value of cardiac magnetic resonance imaging may bring to the diagnosis of the pathophysiology leading to cardiac arrest.

Background

Cardiac arrest is a rare event in the paediatric population with a wide differential diagnosis, including hypertrophic cardiomyopathy, anomalous coronary artery origin, myocarditis and arrhythmogenic right ventricular dysplasia (ARVD). This case illustrates not only the importance of prompt bystander cardiopulmonary resuscitation (CPR) and use of an automatic external defibrillator (AED) for good patient outcome but also highlights the value added when cardiac MRI (CMR) complements traditional, first-line echocardiographic imaging for the diagnosis of myocardial infarction due to paradoxical embolus across a previously undetected sinus venosus defect.

Case presentation

A previously healthy 12-year-old girl experienced an unwitnessed loss of consciousness while walking to her family’s car after visiting with her mother at her mother's place of employment. This child was found in the parking lot unresponsive and without a pulse. The mother initiated CPR, while the emergency medical system (EMS) was notified, and an AED retrieved from the daycare centre. EMS arrived quickly and took over resuscitation. The AED did not initially advise a shock. An artificial airway was placed, and then the AED advised a sock, which was delivered. No pulse was appreciated. CPR was resumed, and IV access was obtained. Upon preparing for transport, the child had return of circulation. She remained hemodynamically appropriate at the outside hospital ED with no dysrhythmia and was transferred for further care to the Critical Care Unit at Children's Hospital of Wisconsin (CHW).

Detailed family medical history revealed no history for cardiomyopathies, congenital heart disease, sudden death, early atherosclerosis, or cardiac dysrhythmias. Social history showed her to be in the sixth grade and to enjoy running track with no prior symptoms. Her medical history was significant for a tonsillectomy with adenoidectomy as a youngster. She was on no chronic medications and had no known allergies to medications or other antigens.

The AED records demonstrated pulseless electrical activity (PEA) that persisted throughout her resuscitation. She did have a shock delivered, but she was in PEA at the time and not ventricular tachycardia or ventricular fibrillation. With ongoing resuscitation, she gradually developed a wide complex rhythm that resolved to sinus rhythm with bundle branch block and ST depression.

Upon arrival at CHW, clinical exam showed a sedated, intubated 54 kg with body mass index (BMI) 19.8 kg/m² (67th percentile) preadolescent female with blood pressure 123/71 mm Hg, pulse rate 110 beats/min, temperature 38.1°C (bladder), respiration rate 23 breaths/min and oxygen saturation 97% on FIO2 0.21. She had clear lungs, a regular rhythm and no appreciable murmurs or gallops. There was no hepatosplenomegaly, and distal perfusion was normal.

Investigations

Remarkable labs on admission included troponin I 4.77 ng/mL (elevated, using high-sensitivity troponin assay at CHW, normal range 0.012–0.034 ng/mL), aspartate transaminase 135 IU/L, alanine transaminase 86 IU/L and D-dimer 7.27 µg/mL (increased, normal range <0.5 µg/mL). Urine drug screen was negative. Coagulation data are shown in the (table 1). Admission chest X-ray demonstrated cardiomegaly with left perihilar soft tissue density compatible with atelectasis and bilateral interstitial and alveolar opacities, most pronounced of the right upper and left lower lung fields, possibly reflecting oedema and aspiration changes (figure 1).

Table 1.

Laboratory evaluations

Labs on admission	Additional coagulopathy workup
Platelet count 331×109/L (normal 150–450×109/L)	Factor 5 Leiden - normal
D-dimer 7.27 µg/mL (declined to 1.32 µg/mL on the day of operation)	Protein C activity 133 IU/ dL (normal>66 IU/ dL)
PT 15.4 s (normal12.4–14.6 s)	Protein S activity 134 IU/ dL (normal>75 IU/ dL)
INR 1.19	Lupus anticoagulant not detected
PTT 26.2 s (normal 23.8–35.0 s)	Anticardiolipin IgG<10
TT 17.8 s (normal 13.1–19.8 s)	Anticardiolipin IgM<11

INR, International normalised ratio; PT, Prothrombin time; PTT, Partial thromboplastin time; TT, Thrombin time.

Figure 1.

Admission chest X-ray.

A transthoracic echocardiogram (Echo) showed a large superior sinus venous septal defect with possible partial anomalous pulmonary venous return (PAPVR) of the right upper pulmonary vein to right atrium. Echo demonstrated moderately dilated right atrium and moderate-to-severely dilated right ventricle (RV). Coronary artery origins were normal. There was mildly diminished left ventricular (LV) systolic function. Electrophysiology Service was consulted to aid with management.

CMR was performed to evaluate for causes for her cardiac arrest (ie, occult myocarditis and ARVD). In addition, CMR scan defined the extent of the superior sinus venosus defect, as well as the pulmonary venous drainage. CMR allowed quantification of the shunt volume, ventricular volumes and ventricular systolic function. CMR with myocardial perfusion imaging and assessment for late gadolinium enhancement (LGE) were undertaken using a Siemens 3T Skyra magnet (Siemens Healthcare USA, Malvern, Philadelphia, USA). The large sinus venous defect with PAPVR was demonstrated with pulmonary flow (Qp) to systemic flow (Qs) ratio 2.5:1. The right ventricular (RV) dilation was noted (with RV end-diastolic volume index 180 mL/m² and RV end-systolic volume (ESV) index 69 mL/m²) with preserved RV systolic function (RV ejection fraction (EF)=62%). LV dimensions were normal with normal LV systolic function (LV EF=67%). Coronary artery origins were confirmed as normal. LGE images showed a focal area of transmural LGE of the LV apex, suggesting myocardial infarct (figure 2).

Figure 2.

Cardiac magnetic resonance imaging (Siemens 3T Skyra scanner): (A) axial gradient images demonstrate sinus venosus defect with partial anomalous pulmonary venous return (PAPVR; Qp:Qs=2.5:1). (B) Coronal reformatted image from magnetic resonance angiography shows PAPVR. (C) Two-chamber image shows apical transmural late gadolinium enhancement.

This transmural LGE prompted coronary angiography by catheterisation, where occlusion of the distal left anterior descending (LAD) coronary artery was detected (figure 3).

Figure 3.

Coronary angiogram showing filling defect (arrow) seen for the distal left anterior descending coronary artery, consistent with apical late gadolinium enhancement seen on cardiac MRI scan.

The aetiology of the LGE was thought to be a paradoxical embolus traversing from right to left across the sinus venosus defect. Thus, she had MRI of the brain to look for other embolic phenomenon. Brain MRI showed multiple, small, ill-defined and irregular foci of hyperintensity in the right frontal white matter. Faint restricted diffusion was also seen in these lesions. No significant volume loss or mass effect was seen associated with these lesions. There was also another punctate hyperintense focus in left cerebellum, which also demonstrated faint restricted, judged to likely represent another focus of subacute ischaemic insult. Given the multiple lesions in different vascular territories, radiology service concurred that a central embolic source was the likely aetiology. Lower extremity Doppler US was normal. Haematology was, therefore, consulted to assess for coagulopathy; none was found (table 1).

Treatment

The patient underwent successful operative closure of her sinus venous defect with redirection of the right upper pulmonary vein to the left atrium using a bovine pericardial patch. Transesophageal echo, performed at operation, confirmed the large sinus venosus defect and right upper pulmonary venous drainage (figure 4). She did well and was discharged home 5 days postoperatively on metoprolol succinate, daily furosemide and daily spironolactone; the diuretics were quickly discontinued. She was also on enoxaparin and warfarin initially, as recommended by haematology.

Figure 4.

Intraoperative transesophageal echocardiogram.

Her comprehensive arrhythmia genetic panel demonstrated no pathologic mutation. Thus, despite the very small area of transmural LGE, an automated implantable cardiac defibrillator was not placed because the presumed mechanism of arrest was embolic (not a primary arrhythmia), and she had recovery of normal cardiac function.

Outcome and follow-up

In clinical follow-up 2 years postevent, the patient has done very well with normal neurocognitive testing and with no cardiovascular symptoms, although with brief costochondritis in the first-year post operation. She participates in volleyball and cheer without symptoms and has advanced in school appropriately.

Follow-up Bruce protocol exercise stress testing (at 1 year postoperatively) showed reduced endurance (25th percentile) with no symptoms, normal heart rate and blood pressure response to exercise, no inducible ischaemic changes and no inducible dysrhythmias, although rare junctional and monomorphic, isolated premature ventricular beats at rest. Follow-up echo showed turbulent flow along the right side of the atrial septum, likely related to the orientation of the systemic venous return, but transthoracic Echo could not rule out a residual shunt; hence, she had a transesophageal echo, which confirmed no residual shunt and normal pulmonary venous drainage to the left atrium. Echo continued to show dilation of the right atrium and RV and mild tricuspid regurgitation with estimated RV pressure=39 mm Hg plus RA pressure, good RV systolic function, with estimated LV EF 52%. She is maintained on extended-release metoprolol 50 mg daily.

Discussion

Sinus venosus defect represents~5%–10% of atrial shunts with the superior sinus venosus defect resulting from a deficiency in the wall that normally separates the right pulmonary veins from the superior vena cava (SVC).¹ As such, the SVC seems to override the atria.

Furthermore, a frequent anatomic finding coexisting with a superior sinus venosus defects is PAPVR, especially of the right upper pulmonary vein to the right atrium or directly to the SVC. Inferior sinus venosus defects have also been described, as a defect near the entry point of the inferior vena cava, allowing the inferior vena cava to communicate with both atria. PAPVR is rarely seen with an inferior sinus venosus defect (7% in one series from our institution).² Typically, with a sinus venosus defect, the patient will have a pulmonary flow murmur and widely split second heart sound due to increased blood flow from the left to right shunt. Heart size will appear large on chest X-ray, and pulmonary vascular markings may be increased. Often, though, as child with a sinus venosus defect (even with PAPVR) may be asymptomatic. If diagnosed by transthoracic Echo, during the workup of a systolic pulmonary murmur, though, the sinus venosus defect (and accompanying PAPVR) is typically repaired by operation when the child is 4–6 years of age. This allows appropriate cardiac remodelling so that chronic right heart dilation does not lead to symptoms (especially atrial dysrhythmias) in adulthood. Pathophysiologic consequences of sinus venosus shunt, as well as atrial septal defect, typically beginning in adulthood, may also include paradoxical embolism, cerebral abscess, pulmonary hypertension and right ventricular failure.³

Echo is first-line imaging, but occasionally full definition of all the pulmonary veins cannot be achieved. CMR proves diagnostic. CMR allows anatomic definition, as well as quantification of shunt volume by calculation of pulmonary versus systemic blood flow (Qp:Qs), as is this case. Volumetric quantification is standard CMR practice by planimetry of short-axis ventricular images.⁴

A paradoxical embolus occurs when there is right to left shunt of a microthrombus from the systemic circulation across an atrial shunt. It is a relatively rare, but well-known complication of atrial septal defect (atrial shunt)/patent foramen ovale (PFO).⁵ Paradoxical embolus has been associated with transient ischaemic attack, stroke and cryptogenic abscess and for peripheral emboli.⁶ Typically, the patient has some predisposition to thrombosis—like hypercoagulable state, obesity, sedentary lifestyle, lower body operation (ie, knee or hip replacement operation) or atrial arrhythmia (with thrombi in the appendage). In our patient’s case, no risk factors were identified; she was a healthy preadolescent athlete with normal haematology workup, normal BMI, active lifestyle and unremarkable medical and surgical history with normal lower extremity Doppler ultrasound.

Paradoxical emboli have been reported with sinus venosus defects.⁷ Interestingly, while most cases of paradoxical emboli lead to cerebrovascular insults, myocardial infarct has been noted on rare occasion.^{7 8} For the more common atrial shunt, a PFO, in a systematic study, Kleber et al screened patients with acute myocardial infarction (AMI) and normal coronaries for the presence of PFO. If no other reason for the AMI could be found, it was judged as due to presumed paradoxical embolus. They found 22 (0.45%) of 4848 AMI were likely due to coronary artery embolism in the retrospective arm of their study, and 11 presumed paradoxical coronary artery emboli in 1654 AMI patients studied prospectively (incidence~0.67%).⁶ Shibata et al found a higher incidence of coronary embolism in 2.9% (52) of 1776 adults having de novo acute myocardial infarct, identifying atrial fibrillation as the most common cause, 38/52 (73%).⁹ In a younger cohort, Bartz et al described embolic phenomenon for 45 patients followed at the Mayo Clinic (median age 29 years, 5–34.9 years) with stroke, 30 (67%); transient ischaemic attack, 13 (29%); myocardial infarction, 1 (2%) and renal infarct, 1 (2%). Seventeen patients (38%) had known cardiovascular disease risk factors: tobacco use (10 patients), hypercoagulable states (7 patients), systemic hypertension (3 patients) and hyperlipidemia (2 patients).¹⁰ Thus, even without known cardiovascular disease risk factor, as in this case report, a paradoxical embolus can occur; while rare, this may lead to myocardial infarction.

Learning points.

• First, prompt bystander cardiopulmonary resuscitation and access to an automatic external defibrillator are vital for patient survival with good cognitive outcome.

• Second, sinus venosus defects can be associated not only with right heart dilation but also with paradoxical emboli leading to cerebral and myocardial infarction. While rare, myocardial infarction can be the result of a paradoxical embolus, even without cardiovascular disease risk factors.

• Third, while echo is the first-line diagnostic imaging modality when cardiac arrest occurs, other imaging modalities may provide complementary data that lead to a better understanding of the pathophysiology of the event. Cardiac MRI with perfusion imaging and late gadolinium enhancement aids diagnosis of myocardial diseases, including myocardial infarction, which should be considered after cardiac arrest.