Every paediatric cardiologist can draw the three fetal shunts from memory. What is less often discussed is how the numbers behind those shunts move across gestation, and what that movement tells us about the lesions we see. This piece walks through those numbers, states plainly where each one comes from, and offers a few clinical connections worth thinking about.
A heart already working at full stretch
The fetal heart operates at a combined output several times higher, per kilogram, than the adult heart at rest. In a study of 222 normal fetuses scanned by high resolution Doppler between 13 and 41 weeks, the median combined ventricular output was 425 mL per minute per kilogram of fetal weight, with a median right to left ventricular output ratio of 1.42, confirming the right ventricle as the dominant chamber throughout gestation (Mielke and Benda, 2001).
For comparison, a resting adult produces roughly 75 mL per minute per kilogram, a figure included only to show scale, not as a clinical benchmark. Unusually high output values per kilogram, seen for example in arteriovenous malformations, sacrococcygeal teratoma or placental chorioangioma, have been associated with worse outcomes in small retrospective series, though no single validated threshold exists across the literature.
A myocardium built for rate, not reserve
The fetal myocardium is structurally and functionally immature relative to the adult heart. Filling depends more heavily on atrial contraction throughout gestation, reflected in an E to A ratio that remains below one until late in pregnancy, and calcium handling relies more on calcium entry across the cell membrane than on a mature sarcoplasmic reticulum.
With limited ability to increase stroke volume through the Frank Starling mechanism, heart rate becomes the primary lever for adjusting output, which helps explain why sustained fetal bradyarrhythmias and tachyarrhythmias both carry a real risk of progressing to hydrops.
The mid-gestation handover
The single most useful physiological idea for this audience may be this: the left heart’s main source of preload changes partway through pregnancy. In a Doppler study of 63 normal fetuses, the foramen ovale’s share of combined output fell from 34 percent at 20 weeks to 18 percent at 30 weeks, while the pulmonary circulation’s share rose from 13 percent to 25 percent over the same window, both changes statistically significant and both settling by 30 weeks (Rasanen et al., 1996).
In plain terms, the left ventricle draws proportionally less of its filling from the foramen ovale and proportionally more from the pulmonary veins as the third trimester approaches.
A left heart that cannot accommodate that shift, because of a developing obstruction or hypoplasia, may only reveal the problem once the shift is underway. This connection is offered here as a physiological interpretation, not a stated conclusion of the original paper, and should be presented as a framework rather than a proven mechanism.
The ductus venosus and the liver’s growing claim
The ductus venosus is often taught as carrying roughly half of umbilical venous return, a figure drawn from fetal lamb studies. Human data tell a different story. In 197 low risk pregnancies scanned between 18 and 41 weeks, the fraction of umbilical venous blood shunted through the ductus venosus was 28 to 32 percent at 18 to 20 weeks, falling to 22 percent by 25 weeks and 18 percent by 31 weeks (Kiserud, Rasmussen and Skulstad, 2000).
The human fetal liver claims a growing share of umbilical blood as gestation advances, likely reflecting its own rising metabolic and growth demands, a detail worth remembering when discussing why the older, sheep derived teaching overstates ductal flow in humans.
The same study found that fetuses below the tenth centile for birth weight shunted significantly more blood through the duct than those above the ninetieth centile. This fits the broader pattern of centralisation, where a compromised fetus reroutes a larger share of what it receives directly toward the heart and brain, at the relative expense of the liver.
The placenta as a circuit
The placenta functions as a third vascular bed, sitting in parallel with the fetal body and the fetal lungs, all three drawing from the same combined output. Across 212 low risk pregnancies scanned between 18 and 41 weeks, the placenta received on average 32 percent of combined cardiac output over that window, falling to 21 percent after 32 weeks (Kiserud et al., 2006).
This fall is not because the placenta shrinks. It reflects the fetal body’s continued growth claiming a larger share of a relatively fixed total. In a comparison group of pregnancies with growth restriction, output per kilogram was preserved, but the fraction directed to the placenta fell further still, more so with worsening placental compromise on umbilical artery Doppler.
Because right ventricular output travels largely toward the placenta, placental vascular resistance functions, in effect, as a component of right ventricular afterload. This is a physiological inference rather than a stated finding of the cited paper, but it offers a useful bridge for cardiologists who already think in terms of pressure and volume load.
Oxygen tensions across the circulation
Fetal blood operates at oxygen saturations that would be considered critical after birth. Approximate figures, drawn largely from classic fetal lamb physiology rather than direct human measurement, put umbilical vein saturation near 80 percent, ascending aortic saturation near 65 percent, and descending aortic saturation near 55 percent, with fetal haemoglobin’s lower P50 improving oxygen loading at placental tension.
The pulmonary bed’s late reactivity
The fetal pulmonary vasculature does not behave uniformly throughout gestation. In a randomised study of maternal hyperoxygenation, fetuses between 20 and 26 weeks showed no measurable change in pulmonary blood flow or vascular indices. Fetuses between 31 and 36 weeks, by contrast, showed a clear response, with pulmonary vascular pulsatility indices falling and pulmonary flow rising (Rasanen et al., 1998).
This has a direct clinical application. Maternal hyperoxygenation testing in fetuses with hypoplastic left heart syndrome, used to help predict a restrictive or intact atrial septum, is only meaningful once the pulmonary bed has become reactive, placing a lower practical limit on when the test can be usefully applied.
Bringing the numbers together
| Number | Clinical question it frames |
|---|---|
| Foramen ovale share falls from 34 to 18 percent, 20 to 30 weeks | Why left heart lesions can progress in severity after mid-gestation |
| Pulmonary share rises from 13 to 25 percent, 20 to 30 weeks | Why a restrictive atrial septum becomes more critical later in pregnancy |
| Placenta receives roughly a third of combined output, falling to a fifth near term | Why placental disease is capable of remodelling the fetal heart |
| Ductus venosus shunt falls from around 30 percent to 18 percent by 31 weeks | Why ductus venosus Doppler reflects pressure more than volume |
None of this is new physiology. What changes when the numbers are set out across gestation, rather than as isolated facts, is that the fetal circulation stops reading as a fixed anatomical arrangement and starts reading as a timeline, one along which specific lesions predictably move.
References
- Mielke G, Benda N. Cardiac output and central distribution of blood flow in the human fetus. Circulation. 2001;103:1662 to 1668. DOI: 10.1161/01.cir.103.12.1662
- Rasanen J, Wood DC, Weiner S, Ludomirski A, Huhta JC. Role of the pulmonary circulation in the distribution of human fetal cardiac output. Circulation. 1996;94:1068 to 1073. DOI: 10.1161/01.cir.94.5.1068
- Rasanen J, Wood DC, Debbs RH, Cohen J, Weiner S, Huhta JC. Reactivity of the human fetal pulmonary circulation to maternal hyperoxygenation. Circulation. 1998;97:257 to 262. DOI: 10.1161/01.cir.97.3.257
- Kiserud T, Rasmussen S, Skulstad S. Blood flow and the degree of shunting through the ductus venosus in the human fetus. American Journal of Obstetrics and Gynecology. 2000;182:147 to 153. DOI: 10.1016/s0002-9378(00)70504-7
- Kiserud T, Ebbing C, Kessler J, Rasmussen S. Fetal cardiac output, distribution to the placenta and impact of placental compromise. Ultrasound in Obstetrics and Gynecology. 2006;28:126 to 136. DOI: 10.1002/uog.2832
- Rudolph AM. Congenital Diseases of the Heart: Clinical-Physiological Considerations. 3rd edition. Wiley-Blackwell, 2009. Textbook reference, not indexed on PubMed.
*All quantitative figures were checked against their PubMed abstracts on 23 September 2026. The oxygen saturation and myocardial physiology figures come from Rudolph’s textbook and were not independently reverified against a primary human data source in this pass.