ORIGINAL

ARTICLES
Effects of Patent Ductus Arteriosus on Organ Blood Flow in Infants Born
Very Preterm: A Prospective Study with Serial Echocardiography
Kai-Hsiang Hsu, MD1,2,3, Jimmy Nguyen, MD4,5, Stephanie Dekom, MD5, Rangasamy Ramanathan, MD5,
and Shahab Noori, MD, MS CBTI1

Objective To characterize the effects of a patent ductus arteriosus (PDA) on different organ blood flows in infants
born preterm.
Study design Infants born preterm at £30 weeks of gestational age had daily echocardiography and Doppler as-
sessments of middle cerebral artery, celiac artery, superior mesenteric (SMA), and renal arteries (RA) during the first
postnatal week. Abnormal organ blood flow was defined as either reverse or absent diastolic flow, abnormally low
mean or systolic velocities, or abnormally high pulsatility or resistance index.
Results Twenty-five infants born very preterm (gestational age 27.0
2.1 weeks) were enrolled. PDA presence at
time of measurement increased the risk of abnormal organ blood flows (39% vs 8%, P < .001). Ductal diameter and
left atrium-to-aortic root (LA/Ao) ratio correlated positively with resistance index (celiac artery, SMA, RA), and nega-
tively with mean velocity (ductal diameter: SMA, RA; LA/Ao ratio: RA). A PDA >2.0 mm, LA/Ao ratio >1.4, and their
combination were associated with 8.0 (95% CI 1.6–39.4)-, 6.7 (1.3–34.7)-, and 38.2 (3.2–455.5)-fold increase in risk
of abnormal organ blood flow index, respectively. Abnormal descending aorta flow was detected in only 2% of mea-
surements.
Conclusions Ductal size >2.0 mm and LA/Ao >1.4, especially in combination, are associated with a greater risk of
abnormal organ blood flows. We suggest that Doppler assessment of the renal and superior mesenteric arteries are
more likely to detect systemic hypoperfusion than the descending aorta. (J Pediatr 2019;-:1-6).

P
atent ductus arteriosus (PDA) is common in infants born very preterm and associated with morbidities such as intra-
ventricular hemorrhage, pulmonary hemorrhage, bronchopulmonary dysplasia, necrotizing enterocolitis, and renal
insufficiency.1 Although a PDA with a left-to-right shunt directly increases pulmonary circulation, its effect on organ
blood flows is less clear.2-4 Several studies have shown that PDA can have adverse effects on organ blood flows, including middle
cerebral artery (MCA),5,6 celiac artery (CA),7-9 superior mesenteric artery (SMA),8,10 and renal artery (RA).9,11,12 However,
there is a paucity of information on PDA and echocardiographic characteristics that lead to compromised blood flow to various
organ vascular beds.
Ductal diameter ³1.5 mm13,14 and left atrium to aortic root (LA/Ao) ratio ³1.3-1.513,15,16 are the most commonly used echo-
cardiographic criteria for hemodynamically significant patent ductus arteriosus (hsPDA). Both criteria for hsPDA are devel-
oped based on the degree of clinical symptoms,13 increased left ventricular output to systemic flow,15 or volume overload of
left heart.14,16 However, given the importance of adequate organ perfusion, defining hsPDA based on objective measure of or-
gan blood flows is also important.17 Our objective was to investigate the effects of PDA on various organ blood flows and define
the echocardiographic criteria for hsPDA based on abnormal organ blood flow indices during the first week of postnatal life
among infants born very preterm.

Methods
This prospective study was conducted in the neonatal intensive care unit at Los
Angeles County and University of Southern California Medical Center. Inborn
From the 1Fetal and Neonatal Institute, Division of
infants born preterm at £30 weeks of gestational age were eligible for the study. Neonatology, Children’s Hospital Los Angeles,
Department of Pediatrics, Keck School of Medicine,
University of Southern California, Los Angeles, CA;
2
Division of Neonatology, Department of Pediatrics,
Chang Gung Memorial Hospital Linkou Branch, Taoyuan,
CA Celiac artery MV Mean velocity Taiwan; 3Graduate Institute of Clinical Medical Science,
Chang Gung University, Taoyuan, Taiwan; 4Division of
DOL Day of life PDA Patent ductus arteriosus Neonatology, Cedar-Sinai Medical Center; and 5Division
EDV End-diastolic velocity PI Pulsatility index of Neonatology, Department of Pediatrics, LAC+USC
hsPDA Hemodynamically significant PSV Peak systolic velocity Medical Center, Keck School of Medicine, University of
Southern California, Los Angeles, CA
patent ductus arteriosus RA Renal artery
The authors declare no conflicts of interest.
LA/Ao Left atrium-to-aortic root RI Resistance index
MCA Middle cerebral artery SMA Superior mesenteric artery 0022-3476/$ - see front matter. ª 2019 Elsevier Inc. All rights reserved.
https://doi.org/10.1016/j.jpeds.2019.08.057

1
THE JOURNAL OF PEDIATRICS  www.jpeds.com
Exclusion criteria were chromosomal or major congenital
anomaly, complex congenital heart disease other than small
atrial or ventricular septal defects, and evidence of perinatal
asphyxia. In addition, infants with major complications
including severe intraventricular hemorrhage grade ³3,
spontaneous intestinal perforation, necrotizing enterocolitis,
or culture-proven sepsis within the first week, or deemed to
be too unstable for echocardiography, were excluded. Given
the direct vasoconstricting effect of indomethacin, ultra-
sound data within 6 hours of indomethacin administration
were excluded. Renal insufficiency was defined as an increase
in creatinine by 0.3 mg/dL or by >50% from the baseline.18
Delayed trophic feeding was defined as initiation of feeds af-
ter first week of life, and the treating clinical team was blinded
to echocardiography and vascular Doppler measurements.
The primary outcome was abnormal organ blood flow. The
institutional review board approved this study, and informed
consents were obtained at enrollment.
Transthoracic echocardiography was performed using
Philips IE33 ultrasound machine with a 12 MHz transducer
(GE Healthcare, Chicago, Illinois). Echocardiography and
Doppler studies were performed every day for the first post-
natal week by 1 of the authors. All the analyses were per-
formed off-line by another author who was blinded to
patients’ demographics, clinical outcomes, and the status of
organ blood flow. Echocardiographic assessment included
evaluation of ductal size, ductal flow pattern, maximum sys-
tolic and end-diastolic ductal flow velocity, LA/Ao ratio, end-
diastolic velocity (EDV) of the left pulmonary artery, and the
presence of retrograde diastolic flow in the descending aorta
distal to PDA. Ductal diameter was estimated at the narrow-
est portion, which often was at the pulmonary end of the duc-
tus, using color Doppler mapping. LA/Ao ratio was measured
on M-mode imaging obtained from parasternal long-axis
view.
The same echocardiography machine and probe were used
to assess organ blood flows. We scanned MCA, CA, SMA,
and RA immediately following echocardiography. For each
artery, peak systolic velocity (PSV), EDV, and mean velocity
(MV) were measured, and pulsatility index (PI = [PSV –
EDV]/MV) and resistance index (RI = [PSV – EDV]/PSV)
were calculated. To minimize intrareader variability, the
average of 3 measurements were used. To have an appro-
priate definition for abnormal blood flow index, references
based on infants with comparable gestational and postnatal
age to our population were applied (Table I; available at
www.jpeds.com).19-21 The measured values were considered
abnormal if <2 SD for flow velocity (PSV and MV) or >2
SD for indices (PI and RI) according to the respective
references. In addition, absent or reverse diastolic flow was
considered as abnormal blood flow. For the purpose of
grouping infants to those with and without abnormal
blood flow, we took a conservative approach. Infants were
classified as having abnormal organ blood flow if there
were abnormal blood flow indices in ³2 organs on the
same day of life (DOL) or ³2 abnormal blood flow indices
of the same organ on different DOL.
2 Hsu et al
Volume -  - 2019
Statistics
Statistical analysis was performed using SPSS Statistics 20
(IBM Corp, Armonk, New York) and Stata/IC 15.0 (Sta-
taCorp, College Station, Texas). For statistics, echocar-
diographic and organ blood flow Doppler data were
categorized as DOL 1, 2, 3, 4, and 5-7. For data DOL
5-7, the values were averaged. The characteristics for in-
fants with abnormal organ flow were evaluated with the
Student t or Mann–Whitney U tests for continuous data,
and c2 or Fisher exact tests for categorical data as
appropriate. Correlations between echocardiographic
measures (ductal diameter and LA/Ao ratio) and various
surrogate markers for organ blood flow (PSV, MV, PI,
and RI) were analyzed using linear mixed effects model.
Univariate and multivariate regression analyses were per-
formed to determine predictors of abnormal organ flow.
A receiver operating characteristic curve was used to
determine the PDA diameter and LA/Ao ratio cut-offs
for hsPDA based on abnormal organ blood flow indices.
Mixed-effects logistic regression was used to test the as-
sociation of these cut-offs with abnormal organ blood
flow indices. For all analyses, a P value < .05 was consid-
ered statistically significant.
Results
A total of 25 infants born very preterm with gestational age
27.0
2.1 (mean
SD) weeks and birth weight
947
349 g were enrolled between June 2015 and December
2016. There were 68 neonates born with a gestational age
£30 weeks during the study period. Two parents did not con-
sent to the study. The other reasons for exclusion were timing
of birth when the sonographer was unavailable and clinical
team deeming patient too unstable for study. During the
study period, we did not perform delayed cord clamping or
cord milking at our institution, and no babies received post-
natal steroids in the first week of life in our cohort. Most in-
fants received empiric antibiotics (24/25; generally ampicillin
with gentamicin). The mean initial hematocrit at admission
was 44.2
7.1%, and the lowest hematocrit during the first
week was 38.2
7.5%. Twelve (48%) infants were transfused
with packed red blood cells. Eight (32%) infants had
abnormal organ blood flow (ie, abnormal blood flow indices
in 2 or more organs on the same DOL or 2 or more abnormal
blood flow indices of the same organ on different DOL), and
all of them had PDA in the first week. There was no difference
in demographics or clinical conditions during the first week,
as well as major morbidities after the study period, between
infants with and without abnormal organ blood flow
(Table II).
A total of 125 data sets were collected (25 infants  5 time
points). Twenty-five data sets were excluded (inadequate
ductal imaging, n = 10; incomplete organ blood flow assess-
ment, n = 13; assessment within 6 hours of indomethacin
administration, n = 2). Therefore, 100 data sets were
analyzed. Ductus arteriosus was patent in 62 studies; 53
- 2019 ORIGINAL ARTICLES

in infants with PDA. The presence of PDA was not associated

Table II. Demographics, clinical conditions, and with abnormal MCA flow index in this study.
morbidities of the study population Concurrent inotrope/vasopressor (dopamine) was used in
Abnormal organ blood flow* 6 of 100 data sets at time of echocardiography; 2 of them were
No Yes P associated with abnormal organ flow index at time of mea-
Demographics (n = 17) (n = 8) value surement. The risk for abnormal organ flow index was
Gestational age, wk 27.3
2.1 26.4
2.0 .296 similar between those with and without dopamine.
Birth weight, g 997
370 841
294 .308 The distribution of Doppler indices for different vascular
Male sex 6 (35%) 3 (38%) 1.000
Cesarean delivery 13 (76%) 7 (88%) 1.000
beds is shown in Table III (available at www.jpeds.com).
Apgar at 1 min 4 (4-7) 3 (2-6) .110 MCA flow was the least affected by the presence of a PDA;
Apgar at 5 min 7 (6-8) 7 (5-8) .124 only 1 MCA flow measurement was abnormal, and this
Maternal hypertension 7 (41%) 2 (25%) .661
Maternal chorioamnionitis 2 (12%) 1 (13%) 1.000
coincided with abnormal CA, SMA, and RA measurements.
Premature rupture of 2 (12%) 2 (25%) .570 RA and SMA were most commonly associated with
membrane >18 h abnormal flow measurements in the presence of a PDA,
Antenatal steroid 17 (100%) 8 (100%) 1.000
Clinical status in
especially on DOL 3 or later (Figure 1). Among infants
the first week with PDA, both ductal diameter and LA/Ao directly
Endotracheal 9 (53%) 5 (63%) 1.000 correlated with RI of CA, SMA, and RA and inversely
intubation
Use of surfactant 12 (71%) 8 (100%) .140
correlated with MV of RA (Tables IV and V; available at
Red blood cells 6 (35%) 6 (75%) .097 www.jpeds.com). There were only 2 measurements with
transfusion abnormal descending aorta flow, and both were concurrent
Started trophic feeding 15 (88%) 3 (38%) .017
Intraventricular 4 (24%) 3 (38%) .640
with abnormal RA flow measurement.
hemorrhage† Infants with ³2 abnormal SMA flow measurements more
Serum Cr, mg/dL, DOL 1 0.82
0.23 0.68
0.14 .138 commonly had delayed initiation of trophic feeding (4/6
Serum Cr, mg/dL, DOL 7 0.69
0.12 0.77
0.23 .432
Use of inotropes/vasopressors 2 (12%) 3 (38%) .283
[67%] vs 3/19 [16%], P = .032), and infants with ³2
Use of indomethacin 3 (18%) 3 (38%) .344 abnormal RA flow measurements more commonly had renal
Use of acetaminophen 0 (0%) 2 (25%) .093 insufficiency (3/6 [50%] vs 1/19 [5%], P = .031). There was
Largest ductal diameter, mm 1.60
0.56‡ 2.36
0.27 <.001
Greatest LA/Ao ratio 1.33
0.68 1.63
0.26 .002
no difference in gestational age, birth weight, or initial serum
Exposure to PDA, d 2 (1-7) 7 (7-7) .009 creatinine between infants with and without ³2 abnormal RA
Morbidities after flow measurements. Infants with ³2 abnormal RA flow mea-
the first week
Spontaneous intestinal 1 (6%) 1 (13%) 1.000
surements had larger PDA on DOL 1 (2.3 vs 1.6 mm,
perforation P = .003) and DOL 3 (2.2 vs 1.6 mm, P = .011).
Necrotizing enterocolitis 2 (12%) 1 (13%) 1.000 In univariate regression analysis, ductal diameter, LA/Ao
any stage§
Intraventricular hemorrhage 1 (6%) 1 (13%) 1.000
ratio, and duration of PDA exposure were associated with
any grade abnormal organ blood flow and ductal flow pattern and
Intraventricular hemorrhage 0 (0%) 1 (13%) .320 end-diastolic flow velocity in left pulmonary artery were
grade ³3
Renal insufficiency{ 1 (6%) 3 (38%) .081
not. In multivariate regression analysis, only ductal diameter
and LA/Ao remained significant. Using receiver operating
Cr, creatinine. characteristic curve analysis, ductal diameter 2.0 mm and
Data are mean
SD, median (IQR), or n (%).
*Infants were classified as having abnormal organ blood flow if they had ³2 abnormal flow LA/Ao 1.4 were identified as cutoff values for hsPDA in rela-
indices (see text). tion to abnormal organ blood flow indices (Figure 2;
†No intraventricular hemorrhage grade 3 or 4 in the first week.
‡These data exclude 2 infants with no PDA. available at www.jpeds.com). If ductal diameter >2.0 mm
§None had a surgical necrotizing enterocolitis. or LA/Ao >1.4, there was 8.0 (95% CI 1.6-39.4)- or 6.7
{Renal insufficiency was defined as an increase in creatinine by 0.3 mg or by >50% from the
baseline. (95% CI 1.3-34.7)-fold increase in risk for abnormal organ
blood flow index, respectively. With combined ductal
diameter >2.0 mm and LA/Ao ratio >1.4, the risk of
were purely left-to-right, 9 bidirectional (predominantly left- abnormal organ blood flow index increased to 38.2-fold
to-right), and none was purely right-to-left. There was no (95% CI 3.2-455.5). Distribution of infants with and
difference in proportion of abnormal organ blood flow without abnormal organ blood flow, and their relationship
indices between left-to-right and bidirectional shunts (22/ to ductal diameter, LA/Ao and PDA exposure duration are
53 [42%] vs 3/9 [33%], P = .728). Of studies with a PDA, demonstrated in Figure 3.
40% (25/62) showed at least 2 abnormal organ blood flow in-
dex vs 8% in studies without a PDA (3/38, P < .001). In those
sets in which ductus was closed (n = 38), the cause of the 3 Discussion
abnormal measurements was unclear. Infants with PDA
had a 12.3-fold greater risk for any abnormal organ blood In this prospective study of infants born very preterm, we
flow index (95% CI 2.3-64.2). All abnormal CA and SMA found a PDA to be associated with abnormal systemic organ
flow indices and 91% of abnormal RA flow indices occurred blood flow during the first week of postnatal life.

Effects of Patent Ductus Arteriosus on Organ Blood Flow in Infants Born Very Preterm: A Prospective Study with Serial 3
Echocardiography
THE JOURNAL OF PEDIATRICS  www.jpeds.com
Figure 1. The number and percentage of abnormal blood flow measurements among those with a PDA.
Furthermore, PDA diameter >2.0 mm and LA/Ao ratio >1.4
emerged as the best predictors of abnormal organ blood flow
indices. Therefore, as far as systemic perfusion is concerned,
these measures could be used to define the presence of a he-
modynamic significant PDA.
Although previous studies were not as comprehensive and
typically evaluated only 1 or 2 peripheral vascular beds, they
reported an alteration or reduction in organ blood flow in the
presence of a PDA.5-12 We extended this observation to daily
and simultaneous assessment of 4 different vascular beds.
Furthermore, we defined a threshold for declaring an organ
blood flow as abnormal based on normative data and were
able to describe the pattern of abnormal blood flow for a spe-
cific organ over the first postnatal week (Figure 1). We found
the impact of PDA to be different on various vascular beds.
Flows in the SMA and RA were commonly affected by PDA
and the MCA and CA were seldom affected. Given the
compensatory increase in stroke volume in the context of
preductal supply of brain blood flow and preferential blood
supply to the brain as a function of cerebral autoregulation,
maintenance of MCA flow is not surprising. Similarly,
Shimada et al found preservation of blood flow in anterior
cerebral artery but low flow indices in postductal vascular
beds in the presence of a significant PDA.11 As for the CA,
the low resistance of the large vascular bed of the liver and
spleen dampens the effect of diastolic flow runoff in this
artery and as such, flow indices are less affected compared
with the SMA and RA.9
Although there is no consensus on what constitutes a
hsPDA, its definition has evolved over the last few decades
4 Hsu et al
Volume -
from a clinical to echocardiographic to various scoring sys-
tems.22 We defined hsPDA based on abnormalities of sys-
temic organ blood flow. We found PDA diameter and LA/
Ao ratio to be useful in predicting abnormal organ blood
flow indices. A PDA diameter >2.0 mm and LA/Ao ratio
>1.4 were the best predictive cutoff values, with the com-
bination of these 2 cutoffs being highly predictive of
abnormal systemic organ blood flow indices (Figure 3).
PDA diameter has long been recognized as a marker of
shunt magnitude.15 Given the fact that blood flow is
directly proportional to the fourth power of the vessel
diameter, it is not surprising that ductal diameter would
be an important determinant of systemic hypoperfusion.
As for LA/Ao ratio, it was the first echocardiographic
index to be used to diagnose a significant PDA. The
utility of this index is based on the premise that with a
significant left-to-right shunt through PDA, pulmonary
venous return increases, thereby enlarging LA dimension.
Indexing LA to a structure that is unaltered by the
increased preload (ie, aortic root), allows for
generalizability of the index to various sizes of the
neonatal population. It is important to note that high
LA/Ao ratio also could be the result of congenital heart
disease such as hypoplastic aorta resulting in abnormal
caliber of the aorta root, and therefore congenital heart
disease should be ruled out in such cases. Despite lack of
specificity to PDA and poor sensitivity in the presence of
large PFO, LA/Ao ratio remains one of the most
commonly reported echocardiographic measures. Our
finding of predictive nature of LA/Ao ratio in identifying
- 2019
Figure 3. Relationship between ductal diameter and LA/Ao
ratio with abnormal organ blood flow. The numbers next to the
circles and triangles indicate DOL.
systemic hypoperfusion indicates that this index is still
relevant.
Although descending aorta retrograde flow can be an in-
dicator of high-volume shunt and has been proposed as a
marker of systemic hypoperfusion,23 we found it to be
insufficient to detect abnormal organ blood flows in the
first week. In our study, descending aorta diastolic flow
reversal was noted only on 2 occasions, both of which coin-
cided with abnormal RA flow measurement. Therefore, or-
gan blood flow indices appear to be more sensitive in
identifying systemic hypoperfusion.
Our study has several limitations. As the study was from
a single unit and had a relatively small sample size, further
studies are needed to ensure our findings are generalizable.
We only examined infants born very preterm in the first
week of life, and the impact of a persistent PDA beyond
the first week of life on organ blood flows may be different.
We defined abnormal organ blood flows based on Doppler
flow indices and limited published data on stable infants
born very preterm. However, we took a conservative
approach by requiring at least 2 abnormal measurements
(<2 SD for flow velocities and >2 SD for PI and RI) to
consider an infant as having abnormal blood flow. Another
limitation inherent to echocardiography is inter/intraob-
server variability. We minimized these problems by having
a single observer perform the echocardiography and a single
observer do all the measurements. Finally, given the afore-
mentioned limitations and uncertainty regarding longer-
term outcome, decision for any intervention based on the
Effects of Patent Ductus Arteriosus on Organ Blood Flow in Infants Born Very Preterm: A Prospective Study with Serial
Echocardiography
ORIGINAL ARTICLES
abnormal organ blood flow should take into consideration
other variables as well.
In conclusion, a PDA diameter >2.0 mm or LA/Ao >1.4
was associated with a greater risk of abnormal organ blood
flow. When both were present, the odds of abnormal
blood flow significantly increased. Echocardiographic
assessment of ductal size and LA/Ao combined with
Doppler assessment of SMA and RA are useful in defining
hsPDA and could aid in the evaluation for systemic hypo-
perfusion. n
Submitted for publication Apr 10, 2019; last revision received Aug 22, 2019;
accepted Aug 27, 2019.
Reprint requests: Shahab Noori, MD, MS CBTI, Fetal and Neonatal Institute,
Division of Neonatology, Children’s Hospital Los Angeles, Department of
Pediatrics, Keck School of Medicine, University of Southern California, 4650
Sunset Blvd, Los Angeles, CA 90027. E-mail: snoori@chla.usc.edu
References
1. Noori S. Patent ductus arteriosus in the preterm infant: to treat or not to
treat? J Perinatol 2010;30:S31.
2. Thibeault DW, Emmanouilides GC, Nelson RJ, Lachman RS,
Rosengart RM, Oh W. Patent ductus arteriosus complicating the respi-
ratory distress syndrome in preterm infants. J Pediatr 1975;86:120-6.
3. Clyman RI, Chorne N. Patent ductus arteriosus: evidence for and against
treatment. J Pediatr 2007;150:216-9.
4. Laughon MM, Simmons MA, Bose CL. Patency of the ductus arteriosus
in the premature infant: is it pathologic? Should it be treated? Curr Opin
Pediatr 2004;16:146-51.
5. Behrman RE, Martin CG, Snider AR, Katz SM, Peabody JL, Brady JP.
Abnormal cerebral blood flow patterns in preterm infants with a large
patent ductus arteriosus. J Pediatr 1982;101:587-93.
6. Lipman B, Serwer GA, Brazy JE. Abnorrmal cerebral hemodynamics
in preterm infants with patent ductus arteriosus. Pediatrics 1982;69:
778-81.
7. El-Khuffash A, Higgins M, Walsh K, Molloy EJ. Quantitative assess-
ment of the degree of ductal steal using celiac artery blood flow to
left ventricular output ratio in preterm infants. Neonatology 2008;93:
206-12.
8. Coombs RC, Morgan ME, Durbin GM, Booth IW, McNeish AS. Gut
blood flow velocities in the newborn: effects of patent ductus arteriosus
and parenteral indomethacin. Arch Dis Child 1990;65:1067-71.
9. Wong SN, Lo RN, Hui PW. Abnormal renal and splanchnic arterial
Doppler pattern in premature babies with symptomatic patent ductus
arteriosus. J Ultrasound Med 1990;9:125-30.
10. Maruyama K, Koizumi T, Tomomasa T, Morikawa A. Intestinal blood-
flow velocity in uncomplicated preterm infants during the early neonatal
period. Pediatr Radiol 1999;29:472-7.
11. Shimada S, Kasai T, Konishi M, Fujiwara T. Effects of patent ductus ar-
teriosus on left ventricular output and organ blood flows in preterm in-
fants with respiratory distress syndrome treated with surfactant. J Pediatr
1994;125:270-7.
12. B€omelburg T, Jorch G. Abnormal blood flow patterns in renal arteries of
small preterm infants with patent ductus arteriosus detected by Doppler
ultrasonography. Eur J Pediatr 1989;148:660-4.
13. Kluckow M, Evans N. Early echocardiographic prediction of symptom-
atic patent ductus arteriosus in preterm infants undergoing mechanical
ventilation. J Pediatr 1995;127:774-9.
14. Evans N, Iyer P. Longitudinal changes in the diameter of the ductus ar-
teriosus in ventilated preterm infants: correlation with respiratory out-
comes. Arch Dis Child Fetal Neonatal Ed 1995;72:F156-61.
15. El Hajjar M, Vaksmann G, Rakza T, Kongolo G, Storme L. Severity of the
ductal shunt: a comparison of different markers. Arch Dis Child Fetal
Neonatal Ed 2005;90:F419-22.
5
THE JOURNAL OF PEDIATRICS  www.jpeds.com
16. Iyer P, Evans N. Re-evaluation of the left atrial to aortic root ratio as a
marker of patent ductus arteriosus. Arch Dis Child Fetal Neonatal Ed
1994;70:F112-7.
17. Sehgal A, Coombs P, Tan K, McNamara PJ. Spectral Doppler waveforms
in systemic arteries and physiological significance of a patent ductus ar-
teriosus. J Perinatol 2011;31:150-6.
18. Selewski DT, Charlton JR, Jetton JG, Guillet R, Mhanna MJ, Askenazi DJ.
Neonatal acute kidney injury. Pediatrics 2015;136:e463-73.
19. Pezzati M, Dani C, Biadaioli R, Filippi L, Biagiotti R, Giani T, et al.
Early postnatal Doppler assessment of cerebral blood flow velocity
in healthy preterm and term infants. Dev Med Child Neurol
2007;44:745-52.
6 Hsu et al
Volume -
20. Papacci P, Giannantonio C, Cota F, Latella C, Semeraro CM, Fioretti M,
et al. Neonatal colour Doppler ultrasound study: normal values of
abdominal blood flow velocities in the neonate during the first month
of life. Pediatr Radiol 2009;39:328-35.
21. Yildirim H, Gungor S, Cihangiroglu MM, Aygun AD. Doppler studies in
normal kidneys of preterm and term neonates. J Ultrasound Med
2005;24:623-7.
22. Shepherd J, Noori S. What is a hemodynamically significant PDA in pre-
term infants? Congenit Heart Dis 2019;14:21-6.
23. Groves AM, Kuschel CA, Knight DB, Skinner JR. Does retrograde dia-
stolic flow in the descending aorta signify impaired systemic perfusion
in preterm infants? Pediatr Res 2008;63:89-94.
- 2019 ORIGINAL ARTICLES

Figure 2. Receiver operating characteristic curves for A, ductal diameter and B, LA/Ao ratio in relation to abnormal organ blood
flows.

Table I. Cutoffs for abnormal value used to define

abnormal organ blood flow in this study
Artery PSV, cm/s MV, cm/s PI RI
MCA <8 <4 N/A >0.98
CA <29 <14 >2.31 >0.86
SMA <14 <4 >3.43 >0.90
RA <9 N/A >2.55 >0.91

N/A, not available.

The cutoff values are derived from normative data.19-21

Effects of Patent Ductus Arteriosus on Organ Blood Flow in Infants Born Very Preterm: A Prospective Study with Serial 6.e1
Echocardiography
THE JOURNAL OF PEDIATRICS  www.jpeds.com Volume -

Table III. Flow velocities and indices for MCA, CA, Table V. Correlation between LA/Ao ratio and organ
SMA, and RA blood flows velocities and indices
Artery PSV, cm/s EDV, cm/s MV, cm/s PI RI Artery Beta 95% CI P value
MCA 36
11 8
3 16
7 1.31
0.68 0.95
0.40 MCA
CA 64
23 18
9 35
15 1.55
0.74 0.75
0.11 PSV 0.112 0.013-0.211 .027
SMA 43
13 13
6 17
8 2.33
1.06 0.85
0.12 MV 0.027 0.039 to 0.919 .423
RA 44
14 6
5 18
5 2.35
1.20 0.87
0.11 PI 0.557 0.193 to 1.308 .146
RI 0.064 0.272 to 0.400 .707
CA
PSV 0.406 0.124 to 0.215 .599
MV 0.012 0.120 to 0.096 .828
PI 0.942 0.188-1.696 .014
RI 0.147 0.047-0.248 .004
SMA
PSV 0.054 0.074 to 0.182 .408
MV 0.002 0.064 to 0.688 .947
PI 1.358 0.264-2.452 .015
RI 0.172 0.053-0.292 .005
RA
PSV 0.075 0.056 to 0.206 .263
MV 0.065 0.110 to 0.020 .004
PI 2.214 1.049-3.378 <.001
RI 0.169 0.066-0.271 .001

Table IV. Correlation between PDA diameter and

organ blood flows velocities and indices
Artery Beta 95% CI P value
MCA
PSV 0.029 0.019 to 0.077 .233
MV 0.015 0.015 to 0.045 .321
PI 0.447 0.144-0.749 .004
RI 0.123 0.262 to 0.015 .081
CA
PSV 0.063 0.156 to 0.030 .184
MV 0.042 0.102 to 0.017 .166
PI 0.346 0.004 to 0.695 .052
RI 0.076 0.029-0.122 .001
SMA
PSV 0.048 0.115 to 0.020 .166
MV 0.060 0.095 to 0.025 .001
PI 0.725 0.206-1.244 .006
RI 0.078 0.022-0.133 .006
RA
PSV 0.004 0.064 to 0.056 .903
MV 0.040 0.061 to 0.018 <.001
PI 0.785 0.218-1.353 .007
RI 0.080 0.030-0.130 .002

6.e2 Hsu et al