ORIGINAL ARTICLE

Relationships between Left Heart Chamber Dilatationon Echocardiography and Left-to-Right Ventricle ShuntingQuantified by Cardiac Catheterization in Childrenwith Ventricular Septal Defects

Selman Gokalp • Ayse Guler Eroglu •

Levent Saltik • Bulent Koca

Received: 2 July 2013 / Accepted: 6 November 2013

� Springer Science+Business Media New York 2013

Abstract Left atrium and/or left ventricle dilatation on

echocardiography is considered to be an indication for

closure of ventricular septal defects (VSD). No study has

addressed the accuracy of using dilated left heart chambers

when defining significant left-to-right shunting quantified

by cardiac catheterization in isolated small or moderate

VSDs. In this study, the relation between dilated left heart

chambers, measured by echocardiography, and left-to-right

ventricle shunting, quantified by cardiac catheterization,

was evaluated in patients with isolated VSD. The medical

records of all patients with isolated VSD who had under-

gone catheterization from 1996 to 2010 were examined

retrospectively. Normative data for left heart chambers

adjusted for body weight (BW) and body surface area

(BSA) were used. The pulmonary-to-systemic flow ratio

(Qp:Qs) was calculated by an oximetry technique. A total

of 115 patients (mean age 7.3 ± 5 years) fulfilled the

inclusion criteria. There was a statistically significant dif-

ference in terms of Qp:Qs between the patient groups with

normal and dilated left heart chambers, when adjusted for

BW and BSA (p = 0.001 and p = 0.002, respectively).

But the relationships between Qp:Qs and left heart cham-

ber sizes on echocardiography were not strong enough to

be useful for making surgical decisions, as left heart

chamber dilatation was not significantly associated with

Qp:Qs C 2 (p = 0.349 when adjusted for BW, p = 0.107

when adjusted for BSA). Left heart chamber dilatation was

significantly associated with Qp:Qs C 1.5 only when it was

adjusted for BSA (for BW p = 0.022, for BSA p = 0.006).

As a result, left heart chamber dilatation measured by

echocardiography does not show significant left-to-right

ventricle shunting, as quantified by catheterization. We still

advocate that catheter angiography should be undertaken

when left heart chambers are dilated in echocardiography

in order to make decisions about closing small- to mod-

erate-sized VSD.

Introduction

The medical management of patients with a ventricular

septal defect (VSD) has not changed much over recent

years, but the surgical and interventional management of

these patients has changed significantly. Although the

indications for surgical repair of large VSDs in infancy are

well defined, there is ongoing discussion about the indi-

cations for closure of small- to moderate-sized VSDs. In

older, asymptomatic children with normal pulmonary

artery pressure, it is generally accepted that VSD closure is

indicated if the pulmonary to systemic flow ratio (Qp:Qs)

is [ 2 [23]. However, changes in surgical management and

the development of transcatheter closure methods have

lowered the threshold for closure of small- to moderate-

sized VSDs. Currently, Qp:Qs [ 1.5 or dilatation of the

left atrium (LA), and dilatation of the left ventricle (LV) on

S. Gokalp (&)

Cocuk Kardiyoloji Bilim Dali, Trakya Universitesi Tip

Fakultesi, Balkan Yerleskesi, Edirne, Turkey

e-mail: sgokalp@hotmail.com

S. Gokalp

Division of Pediatric Cardiology, Trakya University Medical

Faculty, Edirne, Turkey

A. Guler Eroglu � L. Saltik

Division of Pediatric Cardiology, Istanbul University Cerrahpasa

Medical Faculty, Istanbul, Turkey

B. Koca

Division of Pediatric Cardiology, Harran University Medical

Faculty, Sanliurfa, Turkey

123

Pediatr Cardiol

DOI 10.1007/s00246-013-0839-5

echocardiography are considered to be indications for VSD

closure [31, 32]. However, no previous studies have

reported on the accuracy of left heart chamber dilatation

measured by echocardiography for predicting significant

LV to right ventricle (RV) shunting as quantified by car-

diac catheterization in patients with isolated small- and

moderate-sized VSDs.

In general, we recommend VSD closure only if

Qp:Qs C 2. All our patients with isolated small- to mod-

erate-sized VSDs who are considered for surgery therefore

undergo cardiac catheterization to determine Qp:Qs, pul-

monary artery pressure, and pulmonary vascular resistance.

The aim of this study was to evaluate the relationships

between LA and LV size measured by echocardiography

and LV-to-RV shunting quantified by cardiac catheteriza-

tion in patients with isolated small- to moderate-sized

VSDs.

Materials and Methods

Patients

The medical records of all children with VSDs who

underwent cardiac catheterization and angiography from

1996 to 2010 were retrospectively examined. Patients with

additional diagnoses that could affect left-to-right shunting

and patients with pulmonary hypertension were excluded.

Pulmonary hypertension was defined as a mean pulmonary

arterial pressure of [25 mmHg on cardiac catheterization.

The main indication for cardiac catheterization and

angiocardiography was LV or LA dilatation on echocar-

diography. Other indications included excessive LV-to-RV

shunting, suspected pulmonary hypertension based on

transthoracic echocardiography findings, and clinical find-

ings that could not be explained solely by the VSD. We

identified 115 patients who met the inclusion criteria.

Informed consent was obtained from each patient’s family

prior to cardiac catheterization and angiography. The study

was approved by the Ethical Committee of Istanbul Uni-

versity Cerrahpasa Medical Faculty.

Echocardiographic Examinations

Transthoracic echocardiography was performed using a

Siemens Acuson CV70 system with a 4-2 (2–4 MHz) or

9-4 (4–9 MHz) transducer, or a General Electric Vivid 3

system with a 3S (1.5–2.6 MHz) or 7S (3–7 MHz) trans-

ducer. Infants were sedated with intranasal midazolam

when necessary. Two-dimensional M-mode, color-flow

Doppler, pulsed Doppler, and continuous-wave Doppler

echocardiography were performed in all patients. Mea-

surements of the LV, LA, interventricular septum, LV

posterior wall, and aorta were performed by M-mode

echocardiography in the long axis view [27]. Normative

data for cardiovascular structures adjusted for body weight

(BW) were used during routine echocardiographic evalu-

ations [11], and normative data adjusted for body surface

area (BSA) were used retrospectively for study purposes

[17]. BSA was calculated using Costeff’s formula [5].

VSDs were classified according to their location and rela-

tionships to the tricuspid annulus and semilunar valves

[30]. Defect size was expressed in terms of the size of the

aortic root. Defects were classified as large if the diameter

of the VSD was at least two-thirds of the diameter of the

aorta, moderate if the diameter of the VSD was one-third to

two-thirds of the diameter of the aorta, and small if the

diameter of the VSD was less than one-third of the diam-

eter of the aorta [23].

Cardiac Catheterization and Angiography

Cardiac catheterization and angiography were performed

under general anesthesia, and local anesthetic was infil-

trated at the puncture sites. Right-sided cardiac catheteri-

zation was performed in all patients to measure the

pressure and oxygen saturation in the pulmonary artery,

superior vena cava, right atrium, and inferior vena cava. In

patients with a patent foramen ovale, the LA, pulmonary

veins, and LV were evaluated in a retrograde fashion. Left-

sided cardiac catheterization was performed in patients

without a patent foramen ovale. The Qp:Qs, pulmonary

vascular resistance, and systemic vascular resistance were

calculated from oximetry and adjusted oxygen consump-

tion parameters according to the age of the patient. A

clinically important shunt was defined as Qp:Qs C 2.

Angiocardiography was performed in the long axial

oblique or four-chamber view depending on the location of

the VSD.

Statistical Analysis

The results are expressed as median, mean ± standard

deviation, or frequency and percentage. The normality of

data was tested using the Shapiro–Wilk test and histo-

grams. Multiple groups were compared using Kruskal–

Wallis one-way analysis of variance. If the Kruskal–Wallis

test showed a significant difference among groups, com-

parisons between groups were performed using post hoc

Mann–Whitney U tests with Bonferroni correction. The

normality of these data was tested using the one-sample

Kolmogorov–Smirnov test. Considering the number and

distribution of patients, a p value of \0.008 was considered

statistically significant. Data analysis was performed using

SPSS statistical software for Windows (Version 17.0; SPSS

Inc., Chicago, IL, USA).

Pediatr Cardiol

123

Results

The 115 patients included in this study were 62 males

(54 %) and 53 females (46 %), with a mean age of

7.3 ± 5 years (range 8 months to 18 years), mean BW of

16 ± 14.3 kg (range 7–73), and mean BSA of

0.87 ± 0.34 m2 (range 0.36–1.83). The locations and sizes

of the VSDs on echocardiographic examination are shown

in Table 1, the 131 concomitant anomalies detected are

shown in Table 2, and the catheter angiography findings

are shown in Table 3.

M-mode echocardiographic measurements adjusted for

BW showed that 25 patients (21.7 %) had normal LA and

LV measurements, 8 (6.9 %) had LA dilatation only, 51

(44.4 %) had LV dilatation only, and 31 (27 %) had both

LA and LV dilatation. The Qp:Qs values were significantly

different among these groups (p = 0.001) (Table 4).

Comparisons between groups showed significant differ-

ences in Qp:Qs between patients with normal LV mea-

surements and patients with LV dilatation (p \ 0.002),

between patients with normal LA and LV measurements

and patients with both LA and LV dilatation (p \ 0.007),

and between patients with LA dilatation and patients with

LV dilatation (p \ 0.008).

M-mode echocardiographic measurements adjusted for

BSA showed that 22 patients (19.1 %) had normal LA and

LV measurements, 20 (17.4 %) had LA dilatation only, 34

(29.6 %) had LV dilatation only, and 39 (33.9 %) had both

LA and LV dilatation. The Qp:Qs values were significantly

different among these groups (p = 0.002) (Table 5).

Comparisons between groups showed significant differ-

ences in Qp:Qs between patients with normal LV mea-

surements and patients with LV dilatation (p \ 0.005) and

between patients with normal LA and LV measurements

and patients with both LA and LV dilatation (p \ 0.001).

When the left heart chamber sizes were classified

according to BW, 2 of 25 patients (8 %) with normal LA

Table 1 Locations and sizes of the ventricular septal defects (VSD)

on echocardiographic examination

VSD type n Small Moderate Total

Muscular

Inlet 3 1 2

Trabecular 6 5 1 18

Outlet 9 7 2

Perimembranous 96 29 67 96

Doubly committed subarterial 1 – 1 1

Total 115 42 73 115

Table 2 Concomitant anomalies in patients with ventricular septal

defects

Pathologic findings N

Ventricular septal aneurysm 80

Aortic valve prolapsus 24

Sub-aortic ridge (non-obstructive) 9

Left persistent superior vena cava 6

Bicuspid aortic valve 4

Right aortic arch 3

Mitral valve prolapsus 3

Double aortic arch 2

Total 131

Table 3 Catheter angiography findings in patients with ventricular

septal defects

Parameter Mean (±SD)

PA systolic pressure (mmHg) 27.13 (7.36)

PA diastolic pressure (mmHg) 11.53 (3.60)

PA mean pressure (mmHg) 17.64 (4.58)

Qp:Qs 1.52 (0.43)

PVR (Wood unit m2) 2.07 (1.17)

SVR (Wood unit m2) 21.49 (8.14)

Total 115

PA pulmonary artery, PVR pulmonary vascular resistance, Qp:Qs

pulmonary to systemic flow ratio, SD standard deviation, SVR sys-

temic vascular resistance

Table 4 Qp:Qs values in patients with normal and dilated left heart

chambers adjusted for body weight

Left heart chambers N (%) Qp:Qs [mean (±SD)]

Normal 25 (21.7) 1.34 (0.35)

LA dilated 8 (6.9) 1.24 (0.26)

LV dilated 51 (44.4) 1.63 (0.48)

LA and LV dilated 31 (27) 1.56 (0.35)

Total 115 (100) 1.52 (0.43)

p = 0.001

LA left atrium, LV left ventricle, Qp:Qs pulmonary to systemic flow

ratio, SD standard deviation

Table 5 Qp:Qs values in patients with normal and dilated left heart

chambers adjusted for body surface area

Left heart chambers N (%) Qp:Qs [mean (±SD)]

Normal 22 (19.1) 1.31 (0.33)

LA dilated 20 (17.4) 1.40 (0.37)

LV dilated 34 (29.6) 1.56 (0.48)

LA and LV dilated 39 (33.9) 1.66 (0.40)

Total 115 (100) 1.52 (0.43)

p = 0.002

LA left atrium, LV left ventricle, Qp:Qs pulmonary to systemic flow

ratio, SD standard deviation

Pediatr Cardiol

123

and LV measurements, 10 of 51 patients (19.6 %) with LV

dilatation only, and 5 of 31 patients (16.1 %) with both LA

and LV dilatation had Qp:Qs C 2 (Table 6). When the left

heart chamber sizes were classified according to BSA, 1 of

22 patients (4.5 %) with normal LA and LV measurements,

2 of 20 patients (10 %) with LA dilatation only, 4 of 34

patients (11.8 %) with LV dilatation only, and 10 of 39

patients (25.6 %) with both LA and LV dilatation had

Qp:Qs C 2 (Table 7). The proportions of patients with

Qp:Qs C 2 did not differ significantly among these groups

classified according to BW (p = 0.349) or BSA

(p = 0.107). Seventy-five of the 90 patients (83.3 %) with

dilatation of one or both left heart chambers adjusted for

BW and 77 of the 93 patients (82.7 %) with dilatation of

one or both left heart chambers adjusted for BSA had

Qp:Qs \ 2.

When the left heart chamber sizes were classified

according to BW, 6 of 25 patients (24 %) with normal LA

and LV measurements, 2 of 8 patients (25 %) with LA

dilatation only, 29 of 51 patients (56.8 %) with LV dila-

tation only, and 17 of 31 patients (54.8 %) with both LA

and LV dilatation had Qp:Qs C 1.5 (Table 8). The pro-

portions of patients with Qp:Qs C 1.5 did not differ sig-

nificantly among these groups classified according to BW

(p = 0.022). Thirty-eight of the 90 patients (42.2 %) with

dilatation of one or both left heart chambers adjusted for

BW had Qp:Qs \ 1.5. When the left heart chamber sizes

were classified according to BSA, 4 of 22 patients (18.1 %)

with normal LA and LV measurements, 8 of 20 patients

(40 %) with LA dilatation only, 17 of 34 patients (50 %)

with LV dilatation only, and 25 of 39 patients (64.1 %)

with both LA and LV dilatation had Qp:Qs C 1.5

(Table 9). The proportions of patients with Qp:Qs C 1.5

were significantly different among these groups classified

according to BSA (p = 0.006). There was a significant

difference in the proportion of patients with Qp:Qs C 1.5

between the group with normal LA and LV measurements

and the group with both LA and LV dilatation (p \ 0.001).

Forty-three of the 93 patients (46.2 %) with dilatation of

one or both left heart chambers adjusted for BSA had

Qp:Qs \ 1.5.

Discussion

In recent years, the risks associated with VSD surgery have

decreased substantially, and the mortality rate is now less

Table 6 Patients with Qp:Qs C 2 and Qp:Qs \ 2 in the groups with

normal and dilated left heart chambers adjusted for body weight

Left heart chambers [n (%)] Qp:Qs C 2 Qp:Qs \ 2 Total

Normal 2 (11.8) 23 (23.5) 25 (21.7)

LA dilated 0 (0) 8 (8.2) 8 (7)

LV dilated 10 (58.8) 41 (41.8) 51 (44.3)

LA and LV dilated 5 (29.4) 26 (26.5) 31 (27)

Total 17 (100) 98 (100) 115 (100)

p = 0.349

LA left atrium, LV left ventricle, Qp:Qs pulmonary to systemic flow

ratio

Table 7 Patients with Qp:Qs C 2 and Qp:Qs \ 2 in the groups with

normal and dilated left heart chambers adjusted for body surface area

Left heart chambers [n (%)] Qp:Qs C 2 Qp:Qs \ 2 Total

Normal 1 (5.9) 21 (21.4) 22 (19.1)

LA dilated 2 (11.8) 18 (18.4) 20 (17.4)

LV dilated 4 (23.6) 30 (30.6) 34 (29.6)

LA and LV dilated 10 (58.7) 29 (29.6) 39 (33.9)

Total 17 (100) 98 (100) 115 (100)

p = 0.107

LA left atrium, LV left ventricle, Qp:Qs pulmonary to systemic flow

ratio

Table 8 Patients with Qp:Qs C 1.5 and Qp:Qs \ 1.5 in the groups

with normal and dilated left heart chambers adjusted for body weight

Left heart

chambers [n (%)]

Qp:Qs C 1.5 Qp:Qs \ 1.5 Total

Normal 6 (11.1) 19 (31.1) 25 (21.7)

LA dilated 2 (3.7) 6 (9.8) 8 (7.0)

LV dilated 29 (53.7) 22 (36.1) 51 (44.3)

LA and LV dilated 17 (31.5) 14 (23.0) 31 (27.0)

Total 54 (100) 61 (100) 115 (100)

p = 0.022

LA left atrium, LV left ventricle, Qp:Qs pulmonary to systemic flow

ratio

Table 9 Patients with Qp:Qs C 1.5 and Qp:Qs \ 1.5 in the groups

with normal and dilated left heart chambers adjusted for body surface

area

Left heart

chambers [n (%)]

Qp:Qs C 1.5 Qp:Qs \ 1.5 Total

Normal 4 (7.4) 18 (29.5) 22 (19.1)

LA dilated 8 (14.8) 12 (19.7) 20 (17.4)

LV dilated 17 (31.5) 17 (27.9) 34 (29.6)

LA and LV dilated 25 (46.3) 14 (23.0) 39 (33.9)

Total 54 (100) 61 (100) 115 (100)

p = 0.006

LA left atrium, LV left ventricle, Qp:Qs pulmonary to systemic flow

ratio

Pediatr Cardiol

123

than 1 % [1, 24]. Additionally, transcatheter VSD closure

using various devices has become increasingly popular

because of the low complication rates and high rates of

successful closure without residual shunting [33]. The

indications for VSD closure have therefore been extended,

and closure is now suggested for small-sized VSDs with

Qp:Qs C 1.5, or even in patients with LA or LV dilatation

on echocardiography [32]. Pediatric cardiology textbooks

still advise that VSD closure is indicated when Qp:Qs [ 2,

but the indications have changed for adults. In asymp-

tomatic patients without pulmonary hypertension who have

left-sided volume overload, many centers recommend VSD

closure with the aim of avoiding late LV dysfunction

secondary to dilatation [7]. The 2008 Guidelines for the

Management of Adults with Congenital Heart Disease

published by the American Heart Association recommend

device closure of a muscular VSD if it is associated with

severe left-sided heart chamber enlargement or hemody-

namically significant left-to-right shunting (Qp:Qs [ 1.5)

[31]. These recommendations will probably be extended to

include children in the future.

Although the mortality rates associated with surgery and

transcatheter closure are very low, these procedures are not

risk free. Several studies reported good long-term out-

comes and low complication rates in patients with small

VSDs who were treated conservatively, and closure should

not considered as the only treatment option [8–10].

In this study, we evaluated the relationships between LA

and LV measurements on echocardiography and LV-to-RV

shunting quantified by cardiac catheterization in patients

with isolated small- to moderate-sized VSDs. To our

knowledge, no previous studies have reported on the rela-

tionships between left heart chamber sizes measured by

echocardiography and LV-to-RV shunting quantified by

cardiac catheterization in patients with isolated VSDs.

Currently, the main indications for cardiac catheteriza-

tion in children with isolated VSDs are estimation of pul-

monary artery pressure and pulmonary vascular resistance

in case of suspected pulmonary hypertension [22].

Although quantification of Qp:Qs is no longer performed in

many centers, we still perform routine cardiac catheteri-

zation to quantify Qp:Qs in patients with small- to mod-

erate-sized VSDs who are being considered for surgery [3,

20]. There has been considerable interest in developing

non-invasive methods of shunt quantification. Radionu-

clide angiography, Doppler and color-flow echocardiogra-

phy, and cardiac magnetic resonance imaging (CMR)

techniques have been evaluated. Maltz and Treves [21]

described a radionuclide angiocardiography technique for

shunt quantification in 1973. However, scintigraphy can

only provide accurate shunt quantification when Qp:Qs is

between 1.2 and 3.0, as it cannot eliminate the recirculation

effect on the lung dilution curve. Other sources of error

include the non-uniformity of the bolus reaching the heart

and sudden changes in pulmonary blood flow due to crying

or irregular respiration [21]. Later improvements resulted

in an increase in the correlation coefficient of the Qp:Qs

obtained by this method compared with the oximetry

method to as high as 0.92 [12]. However, scintigraphy is

expensive and exposes the patient to radiation, and has

therefore never been a popular method of measuring left-

to-right shunting.

Our clinic used pulsed-wave Doppler ultrasonography

for quantification of intracardiac shunts until the mid-

1990s, but discontinued this because it has some pitfalls

and may induce false quantification. This procedure has

several limitations, as follows. (i) In the presence of LV-to-

RA shunting, the measured RV systolic pressure may be

unreliable. (ii) The pressure gradient between the ventricles

may appear to be lower than the actual pressure gradient

when the defect is small or becomes smaller with con-

traction, and the Doppler beam is not parallel to the

bloodstream. (iii) The Doppler gradient may differ from the

peak-to-peak gradient measured by catheter angiography.

(iv) The calculations assume that the shunt is always in one

direction only, which is inaccurate. (v) Depending on

which method is chosen, either the pulmonary artery and

aorta or the tricuspid and mitral orifices are used for

quantification. These orifices are assumed to be circular,

the flow through them is assumed to be laminar, and the

orifice sizes are assumed to be constant throughout the

cardiac cycle, which is inaccurate. (vi) The VSD is

assumed to have a constant shape and size during systole,

which is inaccurate. A concomitant aneurysm also changes

the geometry and flow through the defect during systole.

(vii) Calculations of the valve areas and the Doppler flow

velocity waveforms depend on manual procedures that are

subject to measurement errors and inter-observer variabil-

ity [4, 26, 29]. The proximal isovelocity surface area

method can also be used for shunt quantification, but is

limited by the difficulty in determining the flow borders in

addition to the above-mentioned limitations [19]. Although

echocardiography measurements show some correlations

with catheter angiography measurements, there are also

some differences between the measurements obtained by

these two modalities.

The American Heart Association guidelines suggest

that, in centers with adequate expertise, CMR or computed

tomography might be useful for assessment of the pul-

monary artery, pulmonary veins, and aorta as well as the

anatomical features of unusual VSDs [31]. CMR may also

be useful for shunt quantification in patients with VSDs

[14]. Velocity-encoded, phase-contrast magnetic resonance

imaging (MRI) measurements of shunt magnitude have

been used for shunt quantification. Data from these studies

indicate that MRI can rapidly and accurately quantify shunt

Pediatr Cardiol

123

magnitude in patients with intracardiac left-to-right

shunting, and can reliably differentiate between patients

with Qp:Qs \ 1.5 and those with Qp:Qs C 1.5 [13].

Beerbaum et al. [2] evaluated this method in pediatric

patients and reported that it was fast, safe, and reliable

compared with oximetry, except in patients with atrial

septal defects. Real-time magnetic resonance velocity

mapping was introduced to save time and overcome the

limitations of phase-contrast CMR [18]. Although these

methods are useful and the results are comparable with

those obtained by oximetry, the long acquisition time,

sedation requirements for children, poor quality of cardiac

gated images in patients with arrhythmias, and high costs

limit their widespread use in clinical practice.

An important limitation common to all methods of shunt

quantification is that sedation affects cardiac physiology.

The required level of sedation differs among echocardi-

ography, radionuclide scintigraphy, CMR, and catheter

angiography. CMR may be the most promising of these

methods of shunt quantification, but cardiac MRI is not

available in our clinic. Cardiac catheterization is therefore

still our gold standard diagnostic technique.

Parameters such as the serum B-type natriuretic peptide

(BNP) level may be useful in combination with the Qp:Qs

value for making operative decisions. The serum levels of

BNP and N-terminal proBNP may increase in response to

volume and pressure overloading, or systolic and diastolic

ventricular dysfunction. These levels are useful for the

diagnosis and management of a variety of clinical condi-

tions, especially in adults with congestive heart failure.

Although a number of studies have evaluated BNP levels in

patients with congenital heart disease, the clinical useful-

ness of serum BNP levels in these patients is less clear, and

the reported data are inconsistent [6, 15, 25]. As we did not

measure serum BNP levels routinely before cardiac cath-

eterization, they are available for only a few of our

patients. Even though there is a positive correlation

between the degree of shunting and the BNP level, there is

no defined cut-off value that can be used for making

operative decisions. The serum BNP level is also affected

by many other factors including sex, obesity, renal disease,

and age, which makes its value more questionable.

In this study, we found significant differences in Qp:Qs

between patients with normal and dilated left heart cham-

bers adjusted for BW and BSA. The Qp:Qs values were

significantly different among patients grouped according to

left heart chamber size adjusted for BW and BSA

(p = 0.001, p = 0.002, respectively). This reflects the

increasing sizes of the left heart chambers as the shunt

increases. The Qp:Qs was higher in patients with either LV

or combined LA and LV dilatation on echocardiography

(adjusted for BW) than in patients with normal LA and LV

measurements. It was interesting that the Qp:Qs was lower

in patients with LA dilatation only than in patients with

normal LA and LV measurements. LA dilatation had no

effect on shunt size, whereas LV dilatation was the echo-

cardiographic finding most strongly associated with shunt

size. Jarmakani et al. [16] reported that VSDs have earlier

and more significant effects on LV and LA end-diastolic

volume than on LV mass. They postulated that there is a

strong relationship between increased LV end-diastolic

volume and increased pulmonary artery blood flow,

resulting in an increase in Qp:Qs. Our findings partially

support this theory, as increased shunting results in LV

dilatation, which was associated with increased Qp:Qs.

However, this association was not strong enough to

determine clinically significant shunting based on echo-

cardiography findings. As shunting causes concurrent

dilatation of the LA and LV, it was unexpected that there

was no association between LA dilatation and Qp:Qs. But

LA size is not linearly correlated with weight or BSA, even

in healthy children. There is also substantial overlap in LA

size between healthy children and children with VSDs. It is

therefore accepted that LA size is variable in childhood

[28]. Our finding might reflect natural variations in LA

size.

The relationships between Qp:Qs and left heart chamber

sizes on echocardiography were not strong enough to be

useful for making surgical decisions, as left heart chamber

dilatation was not significantly associated with Qp:Qs C 2

(p = 0.349 when adjusted for BW, p = 0.107 when

adjusted for BSA). However, left heart chamber dilatation

was significantly associated with Qp:Qs C 1.5 (p = 0.022

when for adjusted BW, p = 0.006 when adjusted for BSA).

It is not clear why the results became statistically signifi-

cant when the measurements were adjusted for BSA rather

than BW. This may have occurred because adjustment for

BSA resulted in an increase in the number of patients who

were considered to have LA dilatation, and consequently

an increase in the number of patients with both LA and LV

dilatation, resulting in a significant difference between

patients with normal LA and LV measurements and

patients with both LA and LV dilatation (p \ 0.001).

If we had based our decisions regarding VSD closure on

echocardiography findings, closure would have been con-

sidered unnecessary in nearly three-quarters of these

patients according to the classical indications. According to

measurements adjusted for BW, two patients with normal

LA and LV measurements on echocardiography who had

Qp:Qs C 2 would have been followed clinically instead of

undergoing surgery; and according to measurements

adjusted for BSA, one such patient would not have

undergone surgery. If we had lowered the threshold for

significant shunting to Qp:Qs C 1.5, as some centers do,

VSD closure would have been considered unnecessary in

nearly half of the patients. According to measurements

Pediatr Cardiol

123

adjusted for BW, six patients with normal LA and LV

measurements on echocardiography who had Qp:Qs C 1.5

would have been followed clinically instead of undergoing

surgery; and according to measurements adjusted for BSA,

four such patients would not have undergone surgery.

The results of this study indicate that the associations

between left heart chamber dilatation on echocardiography

and shunt size are not strong enough to be useful for

making decisions regarding VSD closure in patients with

dilated left heart chambers and small- to moderate-sized

VSDs.

Conclusions

Left heart chamber dilatation measured by echocardiogra-

phy is not significantly associated with significant LV-to-

RV shunting quantified by cardiac catheterization. We

advise that catheter angiography should be performed

when left heart chamber dilatation is observed on trans-

thoracic echocardiography before making decisions

regarding closure of small- to moderate-sized VSDs.

Study Limitations

Quantification of Qp:Qs by catheter angiography was

accepted as the gold standard and was not compared with

other methods of shunt quantification such as Doppler

ultrasonography, scintigraphy, or CMR. The effects of left

heart chamber dilatation on biochemical markers of vol-

ume overload such as BNP and N-terminal proBNP levels

were not compared with Qp:Qs values. The Qp:Qs was

calculated using the oxygen consumption method, even

though several alternative methods are available. Calcula-

tions of Qp and Qs have major limitations, but these lim-

itations may be less important if the Qp:Qs value is used.

Ventricular septal aneurysms and aortic valve prolapse are

known to affect the defect size and the degree of left-to-

right shunting. However, as the aim of this study was to

evaluate the relationships between left heart chamber

dilatation detected on echocardiography and Qp:Qs quan-

tified by cardiac catheterization, we do not consider that

these limitations interfere with the validity of our findings.

Conflict of interest The authors have no conflicts of interest to

disclose.

References

1. Backer CL, Winters RC, Zales VR, Takami H, Muster AJ,

Benson DW, Mavroudis C (1993) Restrictive ventricular septal

defect: how small is too small to close? Ann Thorac Surg

56:1014–1018

2. Beerbaum P, Korperich H, Barth P, Esdorn H, Gieseke J, Meyer

H (2001) Non invasive quantification of left-to-right shunt in

pediatric patients. Phase-contrast cine magnetic resonance

imaging compared with invasive oximetry. Circulation

103:2476–2482

3. Carotti A, Marino B, Bevilacqua M, Marcelletti C, Rossi E,

Santoro G, De Simone G, Pasquini L (1997) Primary repair of

isolated ventricular septal defect in infancy guided by echocar-

diography. Am J Cardiol 79:1498–1501

4. Colan SD (2009) Hemodynamic measurements. In: Lai WW,

Mertens LL, Cohen MS, Geva T (eds) Echocardiography in

pediatric and congenital heart disease: from fetus to adult. Wiley-

Blackwell, West Sussex, pp 63–75

5. Costeff H (1966) A simple empirical formula for calculating

approximate surface area in children. Arch Dis Child 41:681–683

6. Cowley CG, Bradley JD, Shaddy RE (2004) B-type natriuretic

peptide levels in congenital heart disease. Pediatr Cardiol

25:336–340

7. Daniel J, Penny G Wesley, Vick III (2011) Ventricular septal

defect. Lancet 377:1103–1112

8. Eroglu AG, Oztunc F, Saltik L, Bakari S, Dedeoglu S, Ahunbay

G (2003) Evolution of ventricular septal defect with special ref-

erence to spontaneous closure rate, subaortic ridge and aortic

valve prolapse. Pediatr Cardiol 24:31–35

9. Gabriel HM, Heger M, Innerhofer P, Zehetgruber M, Mundigler

G, Wimmer M, Maurer G, Baumgartner H (2002) Long term

outcome of patients with ventricular septal defect considered not

to require surgical closure during childhood. J Am Coll Cardiol

39:1066–1071

10. Gersony WM (2001) Natural history and decision-making in

patients with ventricular septal defect. Prog Pediatr Cardiol

14:125–132

11. Henry WL, Ware J, Gardin JM, Hepner SI, McIllay, Weiner M

(1978) Echocardiographic measurements in normal subjects:

growth-related changes that occur between infancy and early

adulthood. Circulation 57:278–285

12. Houser TS, Maclntyre WJ, Cook SA, Go RT, Moodie DS, Ceimo

J, Gallagher JH (1981) Recirculation subtraction for analysis of

left-to-right cardiac shunts: concise communication. J Nucl Med

22:1033–1038

13. Hundley WG, Li HF, Lange RA, Pfeifer DP, Meshack BM,

Willard JE, Landau C, Willett D, Hillis LD, Peshock RM (1995)

Assessment of left-to-right intracardiac shunting by velocity-

encoded, phase-difference magnetic resonance imaging: a com-

parison with oximetric and indicator dilution techniques. Circu-

lation 91:2955–2960

14. Hundley WG, Bluemke DA, Finn JP, Flamm SD, Fogel MA,

Friedrich MG et al (2010) ACCF/ACR/AHA/NASCI/SCMR

2010 expert consensus document on cardiovascular magnetic

resonance: a report of the American College of Cardiology

foundation task force on expert consensus documents. Circulation

121:2462–2508

15. Jan SL, Fu YC, Hwang B, Lin SJ (2012) B-type natriuretic

peptide in children with atrial or ventricular septal defect: a

cardiac catheterization study. Biomarkers 17:166–171

16. Jarmakani MM, Graham TP, Canent RV, Spach M, Capp P

(1969) Effect of site of shunt on left heart-volume characteristics

in children with ventricular septal defect and patent ductus arte-

riosus. Circulation 40:411–418

17. Kampmann C, Wiethoff CM, Wenzel A, Stolz G, Betancor M,

Wippermann CF, Huth RG, Habermehl P, Knuf M, Emscher-

mann T, Stopfkuchen H (2000) Normal values of M mode

echocardiographic measurements of more than 2000 healthy

infants and children in central Europe. Heart 83:667–672

Pediatr Cardiol

123

18. Korperich H, Gieseke J, Barth P, Hoogeveen R, Esdorn H, Pet-

erschroder A, Meyer H, Beerbaum P (2004) Flow volume and

shunt quantification in pediatric congenital heart disease by real-

time magnetic resonance velocity mapping: a validation study.

Circulation 109:1987–1993

19. Kosecik M, Sagin-Saylam G, Unal N, Kir M, Paytoncu S (2007)

Non-invasive assessment of left-to-right shunting in ventricular

septal defects by the proximal isovelocity surface area method on

Doppler colour flow mapping. Can J Cardiol 23:1049–1053

20. Magee AG, Boutin C, McCrindle BW, Smallhorn JE (1998)

Echocardiography and cardiac catheterization in the preoperative

assessment of ventricular septal defect in infancy. Am Heart J

135:907–913

21. Maltz DL, Treves S (1973) Quantitative radionuclide angiocardiog-

raphy: determination of Qp/Qs in children. Circulation 47:1049–1056

22. Mavroudis C, Backer CL, Jacobs JP (2003) Ventricular septal

defect. In: Mavroudis C, Backer CL (eds) Pediatric cardiac sur-

gery, 3rd edn. Mosby, Philadelphia, pp 298–320

23. McDaniel NL, Gutgesell HP (2008) Ventricular septal defects. In:

Allen HD, Driscoll DJ, Shaddy RE, Feltes TF (eds) Moss and

Adams heart disease in infants, children and adolescents includ-

ing the fetus and young adult, 6th edn. Lippincott Williams &

Wilkins, Philadelphia, pp 667–682

24. Nygren A, Sunnegardh J, Berggren H (2000) Preoperative eval-

uation and surgery in isolated ventricular septal defects: a 21 year

perspective. Heart 83:198–204

25. Ozhan H, Albayrak S, Uzun H, Ordu S, Kaya A, Yazici M (2007)

Correlation of plasma B-type natriuretic peptide with shunt

severity in patient with atrial or ventricular septal defect. Pediatr

Cardiol 28:272–275

26. Sabry AF, Reller MD, Silberbach GM, Rice MJ, Sahn DJ (1995)

Comparison of four Doppler echocardiographic methods for

calculating pulmonary-to-systemic shunt flow ratios in patients

with ventricular septal defect. Am J Cardiol 75:611–614

27. Shan DJ, Demaria A, Kisslo J, Weyman A (1978) Recommen-

dation regarding quantitation in M-mod echocardiography:

results of a survey of echocardiographic measurement. Circula-

tion 58:1072–1083

28. Taggart NW, Cetta F, O’Leary PW, Seward JW, Eidem BW

(2010) Left atrial volume in children without heart disease and in

those with ventricular septal defect or patent ductus arteriosus or

hypertrophic cardiomyopathy. Am J Cardiol 106:1500–1504

29. Teien D, Karp K, Wendel M, Human DG, Nanton MA (1991)

Quantification of left to right shunt by echo Doppler cardiography

in patients with ventricular septal defects. Acta Paediatr Scand

80:335–360

30. Tynan M, Anderson RH (2002) Ventricular septal defect. In:

Anderson RH, Baker EJ, Macartney FJ, Rigby ML, Shinebourne

EA, Tynan M (eds) Paediatric cardiology, 2nd edn. Churchill

Livingstone, Edinburgh, pp 983–1014

31. Warnes CA, Williams RG, Bashore TM et al (2008) ACC/AHA

2008 guidelines for the management of adults with congenital

heart disease: a report of the American College of Cardiology/

American Heart Association Task Force on Practice Guidelines

(writing committee to develop guidelines for the management of

adults with congenital heart disease). J Am Coll Cardiol 52:e143–

e263

32. Working Group on Management of Congenital Heart Disease

(2008) Consensus on timing of intervention for common con-

genital heart disease. Indian J Pediatr 45:117–126

33. Zuo J, Xie J, Yi W, Yang J, Zhang J, Li J, Yi D (2010) Results of

transcatheter closure of perimembranous ventricular septal

defect. Am J Cardiol 106:1034–1037

Pediatr Cardiol

123