Pediatric Transthoracic Echocardiography

Pediatric transthoracic echocardiography (TTE) shares its basic tools — 2D imaging, M-mode, color and spectral Doppler — with adult echocardiography, but the underlying logic is different in ways that genuinely matter. Patients range from a premature neonate to a fully grown adolescent; congenital heart disease introduces anatomic possibilities that simply don’t occur in adult practice; and every measurement has to account for a heart that’s still growing. This tutorial covers what makes pediatric TTE distinct: the segmental approach that organizes the whole exam, Z-scores in place of fixed normal ranges, and the display and view conventions that differ from adult practice.

The Segmental Approach

Congenital heart disease presents an enormous variety of anatomic combinations, and describing them consistently requires a systematic framework rather than an ad hoc description. The segmental approach, developed by Van Praagh and colleagues from the 1960s onward (with an alternative nomenclature advocated by Anderson and colleagues also in wide use), treats the heart as a set of building-block segments, each defined by its own intrinsic morphology — never inferred from a neighboring segment.

There are three main segments — veins and atria, ventricles, and great arteries — connected by two segments that act like multidirectional joints between them: the atrioventricular (AV) canal (the AV valves and the atrioventricular septum) and the conus, or infundibulum (the myocardium connecting the ventricles to the great arteries — normally the subpulmonary conus, with the subaortic infundibulum instead being a fibrous continuity between the aortic and mitral valves). This structure is what allows, for example, a left ventricle to be identified by its own morphology (a smooth superior septal surface) rather than by whichever AV valve happens to open into it — an important distinction, since a double-inlet left ventricle or a common AV valve means the “usual” one-valve-per-ventricle assumption doesn’t hold.

The 10-Step Sequential Analysis

A complete segmental analysis works through the anatomy in a fixed sequence:

1. Thoracoabdominal situs — the lateralized arrangement of the abdominal and thoracic organs (spleen, liver, stomach, bronchial branching pattern), described as solitus (normal), inversus (mirror image), or ambiguous (a lack of normal lateralization, associated with heterotaxy syndrome).
2. Cardiac position — levocardia, mesocardia, or dextrocardia, and whether the heart’s position is primary (isolated) or secondary (displaced by an extracardiac process). 3–4. Atrial situs — solitus, inversus, or ambiguous, following the same logic as visceral situs.
3. Ventricular loop — D-loop or L-loop, describing the rightward or leftward looping of the developing ventricular tube.
4. AV alignments/connections — concordant, discordant, straddling, or atretic.
5. Ventriculoarterial alignments — concordant, discordant (as in transposition of the great arteries), or double-outlet right ventricle, among others.
6. Infundibular anatomy — subpulmonary, subaortic, bilateral, or bilaterally absent conus.
7. Relationships between the semilunar valves — their relative spatial orientation to one another.
8. Associated anomalies — described either by hemodynamic importance or in anatomic order from venous entry to arterial exit.

Two principles hold throughout: each segment is described by its own features, not inferred from an adjacent segment, and both a segment’s situs and its connections must be stated explicitly rather than assumed from each other.

In situs solitus, the spleen, stomach, and sigmoid colon are left-sided while the liver and appendix are right-sided; the left lung has two lobes with a longer, hyparterial mainstem bronchus, and the right lung has three lobes with a shorter, eparterial bronchus. Situs ambiguous — typically seen with heterotaxy syndrome — often still allows a “predominant situs” to be assigned from whichever organs remain partially lateralized, even when a clean solitus/inversus label doesn’t fit.

Echocardiographic Z-Scores

Nearly every measurable cardiovascular structure changes size with growth, which means a single fixed “normal range” — the norm in most adult echocardiography — doesn’t work in children. Instead, pediatric echo expresses measurements as Z-scores: the number of standard deviations a measurement falls above or below the mean expected for a child’s body size (usually body surface area or height) or age, with ±2 conventionally marking the threshold of normal.

A few practical points worth knowing:

• Body size predicts structural growth best; age predicts most functional indices best — reflecting the underlying physiologic link between body size, cardiac output, and cardiovascular growth.
• Different published Z-score models can give meaningfully different values for the same measurement in the same patient. Multiple models exist, mostly derived from single-center data, and a large multicenter Pediatric Heart Network model has shown good — but not perfect — correlation with the older models, with the most notable divergence in the smallest structures (e.g., proximal coronary artery diameter).
• Use the same Z-score model consistently for a given patient over time. Switching models between studies can create the appearance of interval change that isn’t real, or mask genuine change.
• Z-scores don’t reliably extrapolate to abnormal populations — premature infants, children with obesity, and children with congenital heart disease all have body-size relationships that may not match the reference population the model was built from. This is particularly true in obesity, where excess adiposity breaks the usual relationship between body surface area and cardiovascular structure size.

Display Orientation and Standard Views

Pediatric TTE uses an anatomically correct display convention: anterior and superior structures appear at the top of the screen, and rightward structures on the left — which, for subcostal and apical views specifically, means the near field (the vertex of the imaging sector, closest to the transducer) is at the bottom of the display. This is a genuinely different convention from many adult imaging protocols, and it matters here because the wide variety of complex anatomy and abnormal cardiac positions in congenital heart disease makes a fixed spatial convention essential for consistent interpretation. The parasternal long-axis view is the one standard exception — by convention, the cardiac apex is displayed on the left of the screen even though it’s a rightward structure in some framings.

Several views are used far more routinely in pediatric TTE than in adult practice, precisely because they answer questions that come up constantly in congenital heart disease but rarely in adult imaging:

• Subcostal coronal (long-axis) — the primary view for determining visceral situs and cardiac position, showing the relative positions of the spine, IVC, and descending aorta at the diaphragm.
• Subcostal sagittal (short-axis) — the bicaval view, showing SVC, intrahepatic IVC, and hepatic veins; also useful for the right upper pulmonary vein and for RV outflow tract obstruction. The atrial septum is roughly perpendicular to the beam here, giving reliable imaging — unlike apical and parasternal views, where the septum runs parallel to the beam and can produce a dropout artifact easily mistaken for a true septal defect.
• Right parasternal views — in dextrocardia, these serve the same role left parasternal views serve in levocardia, with the transducer rotated 90° clockwise relative to the standard left-sided technique so that right-sided structures still display on the left. In levocardia, right parasternal sagittal views are useful for the atrial septum (particularly a superior sinus venosus defect) and for subvalvar, valvar, or supravalvar aortic stenosis Doppler interrogation.
• The “ductal view” (high left parasternal sagittal) — a dedicated long-axis view of a patent ductus arteriosus, the aortic isthmus, and proximal descending aorta, swept right to left from the ascending aorta through the main and proximal left pulmonary artery to the descending aorta — the transition point where a PDA is typically found, usually at a favorable Doppler angle.
• Suprasternal short-axis (“crab view”) — shows all four pulmonary veins draining into the left atrium in one frame, and (by sweeping superiorly and tracking arch branching) determines aortic arch sidedness — a normal left arch’s first branch is the innominate artery, bifurcating into the right carotid and right subclavian arteries. Color mapping along the left innominate vein here can also reveal a left SVC or other systemic venous anomaly.
• Modified/oblique subcostal views — rotating the transducer counterclockwise from the subcostal coronal view (a right anterior oblique view) shows the RV inflow and outflow tracts in one plane, useful for conal septal deviation in tetralogy of Fallot; rotating clockwise to a position between the coronal and sagittal planes (a left anterior oblique view) gives an en face view of the AV valves, particularly valuable for a common AV valve in AV septal defect.

See the Views & Tips for Optimal Imaging tutorial for the standard adult four-window framework this builds on, and Right Ventricle Evaluation and Right Atrium Evaluation for chamber-specific detail relevant to right heart anomalies commonly encountered in this population.

Operational Considerations

• Transducer frequency needs are wider than in adult practice — low-frequency (2–2.5 MHz), high-frequency (≥7.5 MHz), and multifrequency transducers should all be available to span the full range from a preterm neonate to a large adolescent.
• Frame rate matters more than in most adult imaging, since higher pediatric heart rates make temporal resolution a real limitation; reducing depth and sector width increases frame rate, which is useful specifically when assessing valve function.
• Patient positioning is view-specific: supine with knees flexed (relaxes abdominal muscles) for subcostal imaging, left lateral decubitus for apical and left parasternal imaging, right lateral decubitus for right parasternal imaging, and supine with neck extension over a pillow or rolled towel for suprasternal imaging.
• Procedural sedation is often needed under about age 3, or for children unable to cooperate — requiring an institutional policy covering personnel, patient selection criteria, medication dosing, monitoring, and documentation of successes, failures, and adverse events.
• Agitated saline (bubble study) technique and timing differ from typical adult framing: contrast reaching the left atrium and ventricle within three to six cardiac cycles after right atrial opacification suggests an intracardiac (interatrial) shunt; appearance later than six cycles suggests an intrapulmonary shunt instead.

Practical Pearls

• A prominent hepatic vein should never be mistaken for the IVC itself on subcostal sagittal sweeps — a genuinely interrupted intrahepatic IVC instead shows a dilated azygos vein coursing posterior to the descending aorta, with flow directed superiorly.
• On the suprasternal “crab view,” don’t mistake the right middle pulmonary vein for the right upper pulmonary vein, or the left atrial appendage for the left upper pulmonary vein — both are easy substitutions to make at a glance.
• Unlike adult TTE protocols, systemic venous diameters (e.g., IVC size for right atrial pressure estimation) are not routinely measured in pediatric studies, since no pediatric data support a reliable size-to-pressure correlation the way adult data do.
• Parasternal long- and short-axis views in children also routinely characterize proximal coronary artery origin, course, and size — not a standard part of most adult TTE protocols, but essential given the coronary involvement seen in conditions like Kawasaki disease and certain forms of congenital heart disease. See Coronary Arteries for the full approach, including Kawasaki disease.
• In single-ventricle patients following Fontan palliation, deliberate screening for intracardiac thrombus is a genuine part of surveillance imaging — intracardiac thrombus incidence runs 8–13% in this population, reflecting the sluggish flow inherent to the Fontan circulation itself.