Fetal growth of left-sided structures and postnatal surgical outcome in borderline left heart varies by cardiac phenotype

Ultrasound Obstet Gynecol. 2022 May;59(5):642-650. doi: 10.1002/uog.23689.


Objectives: There are two borderline left-heart phenotypes in the fetus: (1) severe aortic stenosis (AS), which is associated with a 'short, fat', globular left ventricle (LV), LV systolic dysfunction and LV growth arrest; and (2) severe left-heart hypoplasia (LHH), which is associated with a 'long, skinny' LV, laminar flow across hypoplastic mitral and aortic valves and arch hypoplasia. Both phenotypes may be counseled for possible single-ventricle palliation. We aimed to compare the rates of left-sided cardiac structure growth and Z-score change over gestation and to describe the postnatal outcomes associated with these two phenotypes. We hypothesized that the left-sided structures would grow faster in fetuses with LHH compared to those with AS, and that those with LHH would be more likely to achieve biventricular circulation.

Methods: This was a retrospective cohort study using data collected in an institutional cardiology database between April 2001 and April 2018. We included fetuses with severe AS or severe LHH, and with at least two fetal echocardiograms. Inclusion criteria for the AS group included: aortic-annulus Z-score < -2.0, severe AS, depressed LV function, retrograde systolic flow in the aortic arch and endocardial fibroelastosis. Inclusion criteria for the LHH group included: aortic-annulus Z-score < -2.0, hypoplastic but apex-forming LV, normal LV function and retrograde systolic flow in the aortic arch. Exclusion criteria were: abnormal cardiac connections, other forms of structural congenital heart disease, cardiomyopathy, history of fetal aortic valvuloplasty and participation in a maternal hyperoxygenation study. Measurements and respective Z-scores for the aortic-valve annulus, mitral-valve annulus, LV long- and short-axis dimensions, along with aortic-arch measurements, were collected longitudinally for each fetus and plotted over time for both cohorts. Mean slopes of change in dimension and Z-scores over gestation were calculated and compared between the two groups using mixed generalized linear regression accounting for repeated measures. A subanalysis was performed, matching six fetuses from each group for initial aortic-annulus Z-score and gestational age, due to the significant differences in these two variables between the original cohorts.

Results: In total, 53 fetuses with 158 echocardiograms were included. In the AS cohort, there were 20 (38%) fetuses with 65 echocardiograms, and in the LHH cohort there were 33 (62%) fetuses with 93 echocardiograms. On the first echocardiogram, LHH fetuses had a later gestational age and a larger aortic-annulus diameter. The rate of aortic-annulus growth was greater in the LHH group compared with the AS group (mean ± SD, 0.15 ± 0.01 mm/week for LHH vs 0.07 ± 0.01 mm/week for AS (P < 0.001)). While the LHH group had a decrease in aortic-annulus Z-score over time, this was at a slower rate than the decrease in the AS group (mean ± SD, -0.04 ± 0.02/week for LHH vs -0.13 ± 0.02/week for AS (P < 0.001)). A similar pattern was seen for the mitral-valve and LV short-axis-dimension Z-scores. Subanalysis of six fetuses from each group matched for initial aortic-annulus Z-score and gestational age demonstrated similar findings, with the LHH group Z-scores decreasing at a slower rate than those in the AS group. Fifty-two of the 53 fetuses were liveborn, one LHH fetus dying in utero. Of the 20 liveborn in the AS cohort, 15 (75%) infants underwent single-ventricle palliation, two (10%) underwent biventricular repair and three (15%) died prior to intervention. Of the 32 liveborn in the LHH cohort, three (9.4%) underwent single-ventricle palliation, 28 (87.5%) achieved biventricular circulation, of which six required no surgery, and one (3.1%) died prior to intervention.

Conclusions: The left-sided cardiac structures grow at a faster rate in fetuses with severe LHH than they do in fetuses with severe AS, and the Z-scores decrease at a slower rate in fetuses with severe LHH than they do in those with severe AS. The majority of infants in the LHH group did not undergo single-ventricle palliation. This information can be useful in counseling families on the expected growth potential of the fetus's aortic valve, mitral valve and LV, depending on the cardiac phenotype. © 2022 International Society of Ultrasound in Obstetrics and Gynecology.

Keywords: HLHC; HLHS; aortic stenosis; fetal echocardiography; hypoplastic left-heart complex; hypoplastic left-heart syndrome; left-heart hypoplasia; single ventricle.

MeSH terms

  • Female
  • Fetal Heart / diagnostic imaging
  • Gestational Age
  • Heart Ventricles
  • Humans
  • Hypoplastic Left Heart Syndrome* / diagnostic imaging
  • Hypoplastic Left Heart Syndrome* / surgery
  • Phenotype
  • Pregnancy
  • Retrospective Studies
  • Treatment Outcome
  • Ultrasonography, Prenatal / methods