September 3rd, 2025 - Left to right unrestrictive PDA example. Here are some metrics when assessing a left to right shunt in order to determine its significance in terms of excess pulmonary blood flow
September 4, 2025 — As we begin a new training year in Neonatal Hemodynamics within the McGill NH-TNE program, one of our core competencies is understanding ductal physiology: what the ductus tells us, how it affects end-organs, and how to temper shunt volume when appropriate. Our local approach is deliberately conservative and physiology-driven: we maintain adequate PEEP, prioritize appropriate nutrition without undue fluid loading (we often target a total fluid intake of ~150 mL/kg/day), and use permissive hypercapnia to avoid alkalosis-induced pulmonary vasodilation. We limit oxygen exposure while keeping saturations in the 90–95% range with close bedside reassessment, and we favor non-invasive support (bCPAP) to reduce ventilator-related lung inflammation that tends to keep the duct open; when a safe window appears, we aim for timely extubation. In infants still intubated beyond 14 days with substantial support needs (FiO₂ >30% and MAP >10) who are not extubation candidates, we commonly use dexamethasone (DART regimen) to facilitate extubation; in our cohorts, the reduction in inflammation has been accompanied by progressive PDA closure in many patients. We avoid the use of diuretics in the early life of these preterm infants. As a comprehensive hemodynamics program, we teach fellows to track the cardiovascular context rigorously, review how echo markers quantify ductal shunting, and follow how these markers evolve over time, always tying signs and Doppler patterns back to first principles. To support this, we have added a concise page reviewing the basic assessment of a large, unrestrictive left-to-right PDA in preterm infants. The key concept is that a large duct tends to equalize pressures across the connection; shunt direction then reflects the relative balance of PVR and SVR. When the duct is large and the flow is left-to-right, PVR is sub-systemic while PA pressure becomes effectively iso-systemic by transmission. With equal pressures, flow divides according to resistance: if SVR exceeds PVR, pulmonary flow rises at the expense of systemic flow, increasing Qp:Qs. The consequent pulmonary venous return loads the left atrium and left ventricle, producing left-sided volume dilation, rising LA pressure, and—once LA capacitance is exceeded—pulmonary edema with tachypnea and retractions. This page walks through the practical TNE views and measurements we use to characterize this physiology at the bedside.
PDA Characteristics
Diameter (mm) (or indexed to LPA diameter or weight)
Flow direction in systole and diastole, time spent left to right if bidirectional
Velocity in systole (mm/s)
Velocity in diastole (mm/s)
Ratio of velocities
Gradients
Indices of Pulmonary Overcirculation
LVO (mL/kg/min); LVO/RVO
LA:Ao ratio
LVEED Z-score; LVEED/Aorta
Pulm Vein d-wave velocity (m/s)
LPA or RPA end diastolic velocity (m/s)
MV E:A ratio
IVRT - Isovolumetric Relaxation Time
Presence of mitral insufficiency, aortic insufficiency
LA-RA mean gradient by PFO
Indices of Systemic Shunt Effect
Flow direction: post-ductal descending aorta
Flow direction: celiac trunk/SMA
Flow direction: MCA/ACA
In PLAX, we can observe that the LA is already subjectively increased in diameter.
Here the LA/Aorta 1.11/0.54=2.1. LA measured at its peak diameter. The LA fills during ventricular systole and empties during ventricular diastole. The Aorta is measured at aortic valve closure (or during Aortic valve closure duration).
Parasternal long axis view with a posterior sweep (modified position on the thorax). One may appreciate the LA and the RA with the inter-atrial shunt. Here we are nicely aligned to visualize the left to right PFO that is streched and that is at high velocity (considering the Nyquist at 104 cm/s). The PFO is restrictive and high velocity. A PW-Doppler should be obtained as the angle of insonnation is well aligned.
In this view, PW-Doppler is obtained at the PFO level, which is left to right, restrictive and of high velocity. Tracing of the contour of the Doppler enveloppe of the PFO outlines a mean gradientat 8.5 mmHg from the LA to RA, indicating higher LA pressure. The peak gradient is 14 mmHg at the peak of atrial filling (when the mitral valve is closed). Values above 5 mmHg or more are concerning of increased LA pressure.
~3 mmHg → v̄ ≈ 0.9 m/s (often physiologic)
5–6 mmHg → v̄ ≈ 1.1–1.2 m/s (concerning ↑LAP)
8 mmHg → v̄ ≈ 1.41 m/s (marked ↑LAP)
10 mmHg → v̄ ≈ 1.58 m/s (high LAP, think LV inflow/ pulmonary venous issues)
Diastolic flow is seen in the LPA by PW-Doppler.
Branch pulmonary artery. We can see by colour that there is the PDA feeding into the PAs. This view is used to assess the PW-Doppler in the RPA and LPA to see if there is flow in diastole due to the PDA shunting.
Diastolic flow is seen in the RPA by PW-Doppler.
PDA is measured at 0.2 cm (2 mm). The ductus arteriosus is measured at its pulmonary end, which is typically the narrowest segment in the “fetal-type” PDA seen in preterm infants. Measurements are taken at end-diastole—aligned with the R wave on ECG—to avoid systolic distension. Use the ductal “tripod” view, and in the same frame measure the left pulmonary artery (LPA) to calculate a PDA:LPA index. (Inner-edge to inner-edge, perpendicular calipers, adequate zoom, and in B-Mode).
Here we can appreciate the diameter of the PDA, as well as its length. It is a 3D structure and its course may be tortuos and should be imaged in multiple planes.
Tripod view of the PDA (PDA-view) outlining that it is about the size of the LPA.
Here we can appreciate the left to right colour flow through the duct and its pulsatility by colour flow.
Here we can appreciate by PW-Doppler that the flow is pulsatile and unrestrictive.
Here we can appreciate by CW-Doppler that the flow is pulsatile and unrestrictive. The peak systolic velocity is 2.44 m/s. The end diastolic velocity is 0.73 m/s. This outlines that the PDA is unrestrictive (ratio of velocities > 2)
Acquire a high left parasternal “ductal cut” (short-axis at the PA bifurcation, sweeping toward the descending aorta), and confirm in a suprasternal sagittal arch view if this provides better alignment.
Use color with a small box and a color scale around ~0.6–0.8 m/s, adjusting gain to outline the lumen/jet without blooming; verify that the jet truly connects the PA to the descending aorta and note shunt direction.
Freeze at end-diastole and measure inner-edge to inner-edge, orthogonal to the ductal axis. Measure at the pulmonary end (the narrowest point in the fetal-type conical PDA typical of preterms); if the duct is tubular or irregular, also record the minimum diameter anywhere along its length and, when helpful, the aortic-end diameter and overall length. In the same cut, measure the LPA diameter to report the PDA/LPA ratio.
For Doppler, align through the duct (often best from suprasternal or subcostal), record Vmax and trace the mean gradient using 4v², and describe the flow pattern (continuous/sawtooth vs pulsatile vs closing) and direction (left-to-right, right-to-left, or bidirectional).
Clip outlining the left to right PDA . We can also appreciate the post-ductal descending aorta by colour doppler, which is an appropriate place to obtain the PW-Doppler in order to evaluate the instantaneous velocities in the descending aorta.
Retrograde flow in the post-ductal descending aorta. Here the retrograde flow is holodiastolic. The suprasternal view or modified suprasternal view are often the best views for alignment with the post- and pre-ductal aorta. It is good practice to Doppler (PW) before and after the duct to ascertain that the retrograde flow in the aorta is truly secondary to the duct and not to other pre-ductal etiologies.
Sweep in short axis outlining first the large atrium and then the dilated LV. In a hemodynamically significant PDA, oxygenated blood continuously “runs off” from the aorta into the pulmonary artery, producing a left-to-right shunt that scales with ductal size, resistance, and the postnatal fall in pulmonary vascular resistance; the resulting rise in pulmonary blood flow returns to the left atrium via the pulmonary veins, increasing LA preload and pressure, which drives LA enlargement and augments transmitral flow (E-wave dominance, shortened deceleration time). This excess preload then loads the left ventricle, which must accommodate a higher diastolic filling volume and generate a larger stroke volume to sustain systemic output despite diastolic runoff, leading to LV volume overload, eccentric remodeling, and dilation; the magnitude of dilation correlates with the Qp:Qs ratio and the presence of continuous (systolic and diastolic) ductal flow.
Here we measure by M-Mode the LV end-diastolic diameter. Z-scores can be calculated based on gestational age. It is also possible to index the LV end-diastolic diameter to the Aorta diameter in M-Mode with values above 2.0 being considered significant for LV dilatation. Here 1.42/0.51 >> 2.0.
LV dilatation subjectively with the cursor line of interrogation along the mid-septal zone up to the middle of the posterior wall to optain the M-Mode to measure the end-diastolic diameter.
As LA pressure rises, pulmonary venous congestion can develop, further amplifying LA enlargement and, in some infants, functional mitral regurgitation that compounds volume load; taken together, the physiology of a PDA is quintessentially a high-flow, high-preload state for the left-sided chambers, explaining the characteristic dilation of the LA and LV. Here we measure the LA/Ao ratio by M-Mode in the short axis. 1.09/0.51=2.13 (>1.5).
Here we can appreciate that despite the high velocity filter on the colour mode at 1 m/s, we are able to visualize pulmonary venous flow. This tells us that the pulmonary venous flow is coming back at a high velocity secondary to torrential QP:QS. High velocities may also be seen in the context of pulmonary vein stenosis, where there's buildup of pressure approximately creating a jet of high velocity through the ostium. At such, it is always important to do a Pulse wave Doppler at the ostium of the pulmonary veins when there is a high velocity profile.
Flow in the pulmonary veins by colour seen with a Nyquist (velocity filter) at 50 cm/s.
PW-Doppler of the right lower pulmonary vein in the crab-view.
On PW-Doppler of the right lower pulmonary vein, a prominent D wave reflects augmented early-diastolic forward flow from the pulmonary veins into the left ventricle. At mitral valve opening, left atrial (LA) pressure briefly exceeds left ventricular (LV) pressure and, coupled with active LV relaxation (“suction”), releases blood that has accumulated in the LA/pulmonary venous reservoir during ventricular systole. The result is a rapid “conduit” emptying into the LV, recorded as a taller, sharper D wave. Situations that increase LA preload or enhance early-diastolic LV suction—such as left-to-right shunting with high pulmonary venous return (e.g., a significant PDA) or tachycardia with shortened diastasis—accentuate the D wave, whereas impaired LV relaxation or elevated LV filling pressures tend to blunt D and may increase the atrial reversal (Ar) wave.
Apical 4 chamber view. Colour box over the mitral valve. Here we can appreciate mitral valve insufficiency. With the left-sided cavities dilatation, you may loose mitral valve coaptation, leading to mitral insufficiency. We can also appreciate here the dilated left atrium and dilated left ventricle.
Increasing LA preload and pressure drives LA enlargement and augments transmitral flow (E-wave dominance, shortened deceleration time).
A5C: here we can observe the inflow in red and the outflow blood in blue going towards the LVOT which is unobstructed. There may be some acceleration at the LVOT, secondary to the increase LV output due to high LV preload from the increase pulmonary venous return, which is due to the high Qp:Qs.
Increasing LA preload and pressure drives a shorter isovolumetric relaxation time since the mitral valve opens earlier during ventricular diastole. As such, IVRT here is 29.9 ms (<30 ms is considered significantly decreased).
The PW-Doppler at the LVOT is used to estimate LV output, which is typically increased in the context of a left to right PDA with a significant trans-ductal volume shunt. LVO of 300 mL/kg/min is considered high and associated with high pulmonary venous return to the LV, leading to increased LV output by Frank-Starkling. In a left-to-right PDA, a portion of aortic output is siphoned into the pulmonary artery. That blood returns to the LA via the pulmonary veins, raising LV preload and LVEDV (Frank-Starling → larger stroke volume). By Doppler, LVO is measured in the LVOT, proximal to the ductus, so it counts all forward flow the LV ejects into the aorta: LVO = Qs (systemic flow) + ductal shunt flow (that will leave the aorta through the PDA and recirculate). Hence LVO rises even if effective systemic flow (Qs) is normal or low. The diastolic “run-off” through the ductus lowers aortic diastolic pressure, which can reduce effective afterload and further facilitate a higher stroke volume. Many infants also have tachycardia from volume load/compensation, which raises minute output (CO = SV × HR).
A2C outlining mitral insufficiency by colour.
3D Echocardiography of the Left Ventricle outlining left ventricular and atrial dilatation (subjective). Volumes can be measured on the 3D volume reconstruction for both the LV and the RV.
This is a subcostal long-axis view where we outline the flow through the patent foramen ovale. Here we can appreciate that the PFO is streched and that there is acceleration / turbulence at its opening due to the high LA pressure relative to the RA pressure. This is an indirect sign of increased Qp:Qs.
This is a subcostal short-axis view where we outline the flow through the patent foramen ovale. Here we can appreciate that the PFO is streched and that there is acceleration / turbulence at its opening due to the high LA pressure relative to the RA pressure. This is an indirect sign of increased Qp:Qs.
Mean gradient is estimated at 5.9 mmHg from the subcostal view of the left to right PFO. The peak gradient is estimated at 14.8 mmHg.
Subcostal view with the colour box over the descending aorta which is lined along the spine. Red flow that is pulsatile that can be seen.
Holodiastolic retrograde flow in the post-ductal aorta from the PW-Doppler obtained in the subcostal view.
PW-Doppler demonstrating that there is holodiastolic retrograde flow in the post-ductal descending aorta, likely secondary to the steal from the duct.
Absent end-diastolic velocities / flow in the celiak artery by PW-Doppler. There is intermittent retrograde flow (one of the cardiac cycles). We typically score it as retrograde if 3 or more cardiac cycles have retrograde flow.
Trans-cranial doppler to obtain the velocities of the middle cerebral artery. Here the Circle of Willis is obtained with a low Nyquiest (15.4 cm/s). The cursor is not perfectly aligned with the flow in the MCA.
Absent end-diastolic flow in the MCA by PW-Doppler
Original description of the Iowa Score in 2020
PDA risk score = (Gestation in weeks × −1.304) + (PDA diameter in mm × 0.781) + (Left ventricular output in mL/kg/min × 0.008) + (maximum PDA velocity in m/s × −1.065) + (left ventricular a' wave in cm/s × −0.470) + 41.
Score ranges between 0 (low risk) and 13 (high risk).
References:
El-Khuffash A, James AT, Corcoran JD, Dicker P, Franklin O, Elsayed YN, Ting JY, Sehgal A, Malikiwi A, Harabor A, Soraisham AS, McNamara PJ. A Patent Ductus Arteriosus Severity Score Predicts Chronic Lung Disease or Death before Discharge. J Pediatr. 2015 Dec;167(6):1354-1361.e2. doi: 10.1016/j.jpeds.2015.09.028. Epub 2015 Oct 21. PMID: 26474706.
Narrowest PDA diameter (mm) by 2D at pulmonary end
Max velocity across PDA ( in m/s);
Tissue Doppler imaging was obtained at apical 4-chamber view. Late diastolic (a′) velocities used at the level of the lateral mitral valve annulus. If the e′ and a′ waves were fused, they measured the single wave as an a′ wave.
El-Khuffash A, Bussmann N, Breatnach CR, Smith A, Tully E, Griffin J, McCallion N, Corcoran JD, Fernandez E, Looi C, Cleary B, Franklin O, McNamara PJ. A Pilot Randomized Controlled Trial of Early Targeted Patent Ductus Arteriosus Treatment Using a Risk Based Severity Score (The PDA RCT). J Pediatr. 2021 Feb;229:127-133. doi: 10.1016/j.jpeds.2020.10.024. Epub 2020 Oct 16. PMID: 33069668.
<29 weeks: echo at 36 to 48 hours of life.
In RCT, infants with risk score of ≥5.0 deemed at high risk for chronic lung disease or death and were randomized