Imaging of Pediatric Cardiac Tumors: A COG Diagnostic Imaging Committee/SPR Oncology Committee White Paper

Abstract

Cardiac tumors in children are rare and the majority are benign. The most common cardiac tumor in children is rhabdomyoma, usually associated with tuberous sclerosis complex. Other benign cardiac masses include fibromas, myxomas, hemangiomas, and teratomas. Primary malignant cardiac tumors are exceedingly rare with the most common pathology being soft tissue sarcomas. This manuscript provides consensus-based imaging recommendations for the evaluation of patients with cardiac tumors at diagnosis and follow-up, including during and after therapy.

Introduction

Cardiac tumors are very rare in children (< 0.4% incidence) and the majority are histologically benign [1, 2]. The most common cardiac tumor in children is rhabdomyoma, comprising 61% of all pediatric cardiac tumors and usually associated with tuberous sclerosis complex (TSC) [3]. Other benign cardiac masses include fibromas (14%), myxomas (8%), hemangiomas (4%), and teratomas (2%) [3]. Malignant cardiac tumors are exceedingly rare and can be primary or secondary. Soft tissue sarcomas comprise the majority of primary tumors with other rare primary malignancies including malignant germ cell tumor, rhabdoid, and primary cardiac lymphoma [4, 5]. In adult patients, cardiac metastases are seen in approximately 10% of malignancies at autopsy [6, 7]. Limited data are available describing the incidence of secondary involvement of the heart in children, but it appears to be less common than in adults.

Presenting signs and symptoms vary depending on tumor type, location, and extent. Cardiac tumors can be detected on routine prenatal ultrasonography, as in the case of rhabdomyomas. Chest pain is more common in children with malignant cardiac tumors [4].

Imaging in Tumor Staging

Imaging is essential for the diagnosis and staging of both benign and malignant cardiac masses. Staging for malignant primary cardiac masses is based upon the histologic type (GRADE: D; SOR 1.92, strong recommendation); e.g. for primary cardiac non-Hodgkin lymphoma (NHL), the International Pediatric NHL Response Criteria [8][9]. No staging criteria have been validated to correlate with survival outcomes in pediatric primary cardiac tumors.

Selection of Imaging Modality

The advantages and disadvantages of each modality for the evaluation of the primary tumor are summarized in Table 1. Availability and degree of expertise in each modality will vary by site. This should be considered in the development of site-specific diagnosis, staging, and surveillance protocols.

Table 1:

Advantages and disadvantages of each modality for the evaluation of the primary tumor

Procedure Name	Timepoint(s)	Advantage(s)	Disadvantage(s)
Transthoracic echocardiography (TTE)	Screening, initial evaluation, follow-up	Excellent spatial resolution Best temporal resolution Evaluates potential associated flow obstruction or valve disease (stenosis and/or regurgitation) Evaluates small (<1cm) lesions, mobile masses, and masses arising from valves *Above advantages further augmented with 3D probe	Restricted field of view (body habitus, poor acoustic windows, significant lung disease) Operator dependent Origin and extent of mass is difficult to delineate Limited tissue characterization Limited imaging of the right heart, mediastinal, and extracardiac structures
Transesophageal echocardiography (TEE)	Initial evaluation when first-line TTE images are suboptimal Intraoperative planning and guidance	Better evaluates atrial and atrioventricular valve masses (due to proximity of the probe to atrial chambers)	Requires deep sedation and/or anesthesia with intubation in pediatric patients
Cardiac Magnetic Resonance (CMR)	Diagnosis/staging, follow-up, surveillance	Superior tissue contrast High temporal resolution Defines likely tissue type, location, size, relationship to adjacent structures and hemodynamic effects Multiplanar	Sedation/anesthesia often necessary >6 months < 6 years age Less availability in centers that lack technical expertise in pediatric CMR
CT	Used when contraindication to CMR exists Select instances to evaluate for calcification and/or coronary artery involvement	Fast May be more readily available than CMR in many institutions Multiplanar High spatial resolution	Timing of contrast to delineate a specific chamber requires expertise especially in smaller patients with faster heart rates Limited soft-tissue contrast Decreased temporal resolution compared to TTE and CMR
PET	Evaluate metastatic disease based on tumor type	Some studies suggest metabolic activity helps predict benign versus malignant lesions Tumor staging	Less overall experience in the literature relative to other modalities

Imaging at Diagnosis

Imaging evaluation of pediatric cardiac masses begins with transthoracic echocardiography (TTE) (GRADE: D; SOR 1, very strong recommendation). Transesophageal echocardiography (TEE) is utilized when TTE imaging is poor or intra-operative evaluation is required. Additional cross-sectional imaging for locoregional staging is not always required depending on the tumor type. For example, no additional imaging is performed in the case of presumed rhabdomyoma(s) in infants.

When additional imaging is needed for the evaluation of the primary tumor, cardiovascular magnetic resonance (CMR) is nearly always the modality of choice (GRADE: B; SOR 1.33, very strong recommendation). CMR imaging can narrow the differential, primarily based on tissue characterization and enhancement profile, as well as define mass location. Computed tomography (CT) can be used to assess for calcifications, for coronary artery involvement, or if a patient cannot tolerate or undergo CMR. In cases of suspected malignancy, additional imaging at diagnosis may include routine chest CT or positron emission tomography (PET)/CT to evaluate for the local invasion of the primary tumor and/or metastatic disease. PET/CT has a limited role in the initial evaluation of primary cardiac masses.

Imaging Parameters

Echocardiography should include a comprehensive study of the mass, including assessment of mass morphology and size, involvement of adjacent cardiac structures, and valvular and ventricular function. Recommended imaging parameters for CMR and CT are summarized in Tables 4–5.

Table 4.

Suggested MRI protocol for evaluation of cardiac lesions. MRI with and without contrast is performed at initial evaluation for all masses not definitively characterized by echocardiography. Follow-up studies can be performed without contrast if studies are checked in real-time without change in size.

Pulse sequence	Coverage & Plane	Contrast phase	Sequence parameters	Required/Optional	Comment
Cine steady-state free precession	Heart long axis, short axis stack, multiple slices through mass for full characterization	Pre-contrast	ST: ≤ 8mm; dX: 1.5x1.5 mm dT: <40 ms/phase	Required
T1-weighted black blood	Short axis or long axis slices through the mass	Pre-contrast	ST: <8mm; dX: 1.5x1.5 mm HR < 90 → diastole HR > 90 → systole	Optional
Fat-suppressed T1-weighted black blood	Short axis or long axis slices through the mass	Pre-contrast	ST: <8mm; dX: 1.5x1.5 mm HR < 90 → diastole HR > 90 → systole	Optional	pre-contrast T1 and T1FS can be useful if suspected teratoma or lipomas
Fat-suppressed T2-weighted black blood	Short axis or long axis slices through the mass	Pre-contrast	ST: <8mm; dX: 1.5x1.5 mm HR < 90 → diastole HR > 90 → systole	Required
First-pass perfusion	1-3 Short axis or long axis slices through the mass (depending on heart rate)	(0.1 mmol/kg of GBCA at 2-4 ml/s, followed by 0.05 mmol/kg of GBCA at 1 ml/s)	ST: <10mm; dX: 2x2 mm <600 ms / acquisition	Required
Fat-suppressed T1-weighted black blood	Short axis or long axis slices through the mass	Post-contrast	ST: <8mm; dX: 1.5x1.5 mm HR < 90 → diastole HR > 90 → systole	Optional
Late gadolinium enhancement	Long axis slices and short-axis stack; add slices to optimize imaging of mass	Post-contrast	ST: <8mm; dX: 1.7x1.7 mm Minimize acquisition duration per R-R interval HR < 90 → diastole HR > 90 → systole	Required	PSIR segmented or single-shot acquisition with optimal TI to null myocardium
Late gadolinium enhancement	Single slice through mass	Post-contrast	ST: <8mm; dX: 1.7x1.7 mm Minimize acquisition duration per R-R interval HR < 90 → diastole HR > 90 → systole	Required	PSIR segmented or single-shot acquisition with long TI (>600 ms)
Triple IR	SA stack or 4-ch stack (can have flow-related artifacts)			Optional

ST = slice thickness; dX = pixel dimension; dT = temporal resolution; HR = heart rate (in beats per minute); GBCA - gadolinium based contrast agent; TI = inversion time; FS = fat-saturated

Table 5.

Suggested CT protocols for evaluation of cardiac masses. All CT protocols require a minimum acceptable resolution of 0.625 mm slice thickness. Reformats includes standard MPRs in axial, coronal & sag in soft tissue windows and lung kernels.

Name	Primary Goal	Coverage	Contrast phase	Gating & Breathing	Comment
Routine Chest	Evaluate mediastinal structures Evaluate for lung mets	From thoracic inlet through upper abdomen	Routine timing for chest (depends on patient size)	Most exams can be done free breathing on newer generation CT scanners with high pitch, low-dose helical scan modes. Gating is optional; ECG-triggered gating can be helpful if available	Additional maximum intensity projection reconstructions (vascular anatomy, pulmonary nodules), minimum intensity projection (airway anatomy), and volume rendering (3-dimensional reconstructions, bony anatomy) may be helpful for specific cases
Cardiac CT Angiography	Evaluate coronary arteries	Pulmonary arteries through bottom of heart	Systemic arterial (aortic)	HR & Institution-dependent	At higher heart rates, systolic phase imaging should be considered in high-pitch helical and prospective ECG-gated modes as well as when setting dose modulation windows in retrospective ECG-gated mode. [40]
Cardiac CT Morphology	Evaluate cardiac mass	Pulmonary arteries through bottom of heart (dependent on mass location)	Approach to contrast timing depends upon location of the mass with the goal to time contrast for the chamber of interest.	Usually retrospective ECG-gating	Performed if CMR is contraindicated If concern for calcification, consider dual-energy scan or non-contrast scan prior to contrast administration

Imaging at Follow-Up

Once a diagnosis has been made, the imaging modality at subsequent time points will be directed by etiology and management. TTE is used to evaluate changes in mass morphology, as well as associated ventricular or valvular dysfunction (GRADE: D; SOR 1.58, strong recommendation). CMR and/or CT may be used to reevaluate larger masses that extend beyond sonographic windows and/or that involve other intrathoracic structures (GRADE: B; SOR 1.25, very strong recommendation). Imaging parameters are unchanged from baseline imaging, except that CMR protocols do not require contrast evaluation in many cases (GRADE: D; SOR 2, moderate recommendation).

Tumor Response Assessment

For benign cardiac masses (Table 2), once the clinical team has determined no further medical or interventional therapy is required, the purpose of imaging is to provide periodic surveillance of the mass(es), and can usually be accomplished by TTE (GRADE: B; SOR 1.33, very strong recommendation). Repeated hemodynamic assessment with CMR and/or CT may be useful in evaluating mass effect on adjacent structures as the child grows if this cannot be achieved by TTE alone (GRADE: D; SOR 1.42. very strong recommendation). Most benign cardiac masses either regress or do not change substantially in size to warrant extensive re-evaluation.

Table 2.

Benign tumors usually managed by surveillance or surgical resection

Tumor Type	Related syndrome	Required surveillance	References
Rhabdomyoma	TSC	• Follow-up echocardiography is not recommended in the first year of life in absence of clinical symptoms    • TTE every 1-3 years in asymptomatic pediatric patients until regression of cardiac rhabdomyomas is documented	[33–35]
Fibroma	Nevoid basal cell carcinoma syndrome (i.e., Gorlin-Goltz syndrome)	• Baseline TTE should be performed at the time of diagnosis of GS, ideally within the first 6 months of life	[36]
Teratoma	n/a	• Routine TTE for evidence of recurrence, possibly because of incomplete surgical resections, and possible malignant transformation	[37]
Myxoma	Carney Complex	• All myxomas: After surgery, TTE should document complete tumor resection and follow-up echocardiograms are recommended every 6 months for a minimum of 4 years.    • Carney Complex: TTE should start during the first 6 months of life and be performed at least once a year thereafter.	[38]
Hemangioma	n/a	• Periodic TTE is recommended, considering the potential for tumor growth or recurrence after surgery	[39]

For malignant tumors (Table 3), CMR may be preferred for re-evaluation of cardiac involvement. CT might be considered for the reassessment of local invasion and/or metastatic disease (GRADE: D; SOR 1.75, strong recommendation). Although response assessment criteria specifically for cardiac tumors have not been evaluated, the RECIST v1.1 is the most commonly used response assessment tool for children with sarcomas [9, 10]. Accurate assessment of size and response to therapy is best accomplished by estimating tumor volume (mL), given the irregularity of most masses and challenges with repeated two-dimensional comparisons [11]. Because children have a wide range in body sizes, indexing tumor volume relative to body surface area provides a clinically useful reference for the size of the tumor relative to the size of the thoracic cavity; this is particularly relevant for serial follow-up of benign tumors such as fibromas [11]. In cases of malignant tumors with cardiac metastases, direct measurements of size/volume may be limited by the overall extent of disease, including multiple masses. In these cases, the response may be evaluated by the secondary effect on cardiac and pulmonary hemodynamics (e.g., compression of the heart, airways, and/or vascular structures).

Table 3.

Primary & Secondary malignant cardiac tumors and surveillance recommendations

Tumor type	Subtypes	Tumor response assessment and other considerations	Reference
Sarcomas	Rhabdomyosarcoma Angiosarcoma Malignant fibrous histiocytoma Fibrosarcoma Undifferentiated sarcoma	CMR or CT every 3 months in the first year, every 4 months in the second and third years, and then yearly after treatment	[10]
Lymphoma		Surveillance per lymphoma	[8]
Secondary	Wilms tumor Hepatoblastoma Neuroblastoma Ewing sarcoma Osteosarcoma Medulloblastoma	Tumor thrombus extending into the right atrium from abdominal solid tumor may be surgically removed. Continued surveillance per primary malignancy recommendation. Rare metastatic lesions from Ewing, OS, MB; surveillance per primary malignancy

Imaging Off Therapy/Surveillance

Long-term surveillance imaging of cardiac tumors in children should be primarily performed with TTE and/or CMR (GRADE: D; SOR 1.42, very strong recommendation). The goals of surveillance include monitoring histologically benign tumors for growth or further cardiac compromise or monitoring for recurrence in patients after benign tumor resection or completion of definitive therapy for malignant tumors. Surveillance should not be performed in patients where it will not impact or guide prognostication or therapeutic decisions (GRADE: D; SOR 1.33, very strong recommendation). The frequency and modality of surveillance are dependent on tumor type, size, location, and degree of cardiac involvement. Consider more frequent assessment in tumors with high recurrence potential (e.g., myxomas), valvar or hemodynamic consequence, or myocardial impairment. In patients with a hemodynamic impact from the mass, surveillance by TTE and/or CMR may be warranted every few months. More typically, TTE can be performed on a yearly basis and CMR less frequently in benign tumors with lower potential for recurrence and less cardiac involvement (GRADE: D; SOR 2.08, moderate recommendation). Some benign cardiac masses are associated with genetic syndromes and may warrant additional surveillance (Table 2). [1]

Are there late effects that change the goal of surveillance imaging?

Childhood cancer survivors (CCS) have a 15-fold increased risk of developing heart failure (HF) compared to siblings [12]. This risk has a strong dose-dependent relationship with cumulative anthracycline exposure. Other risk factors include younger age at treatment, exposure to chest radiation, and hematopoietic cell transplantation (HCT). To facilitate the early identification of asymptomatic cardiomyopathy, screening recommendations endorse the use of TTE to detect a subclinical reduction in the left ventricular ejection fraction (LVEF) in at-risk survivors (GRADE: A; SOR 1.17, very strong recommendation) [13]. Increasingly, CMR is used clinically to provide more accurate measurements of ejection fraction and simultaneous evaluation of the myocardium for evidence of injury or inflammation (GRADE: A; SOR 1.08, very strong recommendation) [14, 15]

Novel and future imaging techniques

Multiple quantitative imaging techniques are emerging that may allow for improved cardiac mass characterization. T1 and T2 mapping are becoming routine CMR sequences that provide quantitative tissue relaxation times. Mapping of cardiac masses may improve the interpretation of tumor composition beyond using subjective or relative assessments of signal intensity [16–19]. Automated quantitative CMR perfusion techniques are translating to clinical practice for the evaluation of myocardial ischemia [20–22]. In time, it is expected these techniques will be applied to tumor characterization and may complement qualitative interpretations of tumor vascularity. Electrocardiographically-gated dual-energy CT has recently become an available imaging option that allows for quantitative measurements of iodine uptake within focal cardiac lesions. Iodine uptake by dual-energy CT is superior to the use of Hounsfield Unit measurements to discriminate cardiac thrombi from tumors, albeit performance remains inferior to CMR for this purpose [23, 24]. Diffusion-weighted MRI (DWI) is commonly used for the assessment of masses elsewhere in the body. Although DWI is limited in the heart due to signal loss from cardiac motion, some investigators have found value in the sequence for detection and risk stratification of cardiac lesions [25, 26]. Ultrasound-enhancing agents can be used to assess the vascularity of the mass and to exclude thrombus (without internal blood flow, contrary to malignant tumors with avid angiogenesis) [27–29]. Finally, radiomic techniques are now being applied to both CT and MR imaging data of cardiac masses, which may provide additional complementary diagnostic information in the future [30–32]. Continued development of such emerging techniques offers the potential to improve diagnostic performance and allow for more sophisticated non-invasive risk stratification for children with a cardiac mass.

Abbreviations

US

Ultrasound

CT

Computed tomography

PET

Positron emission tomography

CMR

Cardiovascular magnetic resonance

TTE

Transthoracic echocardiography

TEE

Transesophageal echocardiography

LVEF

Left ventricular ejection fraction

RECIST

Response Evaluation Criteria in Solid Tumors

DWI

Diffusion-weighted imaging