Original Article

Clinical utility of the ventricular septal defect diameter to aorta root

diameter ratio to predict early childhood developmental defects or lung
infections in patients with perimembranous ventricular septal defect
Jin-Xiang Liu*, Jing-Hua Wang*, Si-Rui Yang, Min Liu, Yang Xu, Jing-Hui Sun, Chao-Ying Yan
Institute of Pediatrics, First Affiliated Bethune Hospital, Jilin University, Changchun 130021, China

ABSTRACT Background: Ventricular septal defect (VSD) is the most frequent type of congenital heart disease. Conventional methods
to evaluate VSD size and severity are both invasive and cumbersome to perform. We investigated whether the ratio between
the diameter of the defect and the aortic root diameter (DVSD/DAR) would accurately reflect the degree of shunted blood and
the severity of VSD in children with perimembranous VSD.
Methods: We recruited 987 children with perimembranous VSD (pmVSD) and used color Doppler echocardiography to
calculate DVSD/DAR. 987 healthy children were recruited as control group. The pmVSD group was further stratified into four groups
according to age (1 to 4 y) and again into four groups according to the DVSD/DAR ratio: children whose DVSD/DAR was 1/5 to <1/4,
1/4 to <1/3, 1/3 to 1/5, or 1/2 to <2/3 were assigned to groups A, B, C, and D, respectively. Height, weight, infection scores and
systemic-pulmonary circulation ratio (QP/QS ratio) were compared among groups A, B, C and D. Then the relationship between
the DVSD/DAR ratio and height, weight, QP/QS ratio, infection score were analysed by linear regression analysis.
Results: Compared to age-matched children without VSD (the controls), the mean height and weight of children in the
pmVSD group were lower, and heights and weights were negatively correlated with the DVSD/DAR ratio. This ratio was
significantly reduced in groups C and D compared to control group (both P<0.05). Infection scores of groups A and B were
significantly higher only in the one-year-old subgroup, but were significantly higher in groups C and D for all ages compared to
the control group (both P<0.05). QP/QS ratio of group C and D were higher than group A and group B (all P<0.05). Moreover,
QP/QS ratio of group D for all ages were more than 2. In addition, linear regression analysis revealed that the DVSD/DAR ratio
negatively correlated with height and weight and positively correlated with the QP/QS ratio and infection score.
Conclusions: Our results suggest that the DVSD/DAR ratio accurately reflects the growth and pulmonary infection rates in
children with pmVSD. This ratio, therefore, may be a useful additional reference index to predict the consequences of pmVSD.
KEYWORDS Perimembranous ventricular septal defect (VSD); growth; pulmonary infection; ratio between defect diameter and the
aortic root diameter (DVSD/DAR)

J Thorac Dis 2013;5(5):600-604. doi: 10.3978/j.issn.2072-1439.2013.09.05

Introduction spontaneous postnatal closure by the time the child reaches

5 years of age (2), the outcomes are usually growth retardation
Ventricular septal defect (VSD) is the most common type of and recurrent infections (3). It is very important, accordingly,
for pediatric cardiologists to predict which complications will
congenital heart disease (1). While some VSDs can undergo
occur and to choose the most suitable time for treatment. This
will result in achieving more scientific-based and individualized
*These authors contributed equally to this work.
therapy in children with VSD.
Corresponding to: Chao-Ying Yan, MD, PhD. Institute of Pediatrics, First
Affiliated Bethune Hospital, Jilin University, Changchun 130021, China. Email:
Previous studies found that VSD diameter closely correlated
[email protected]. with the different outcomes. However, using simple VSD
diameter to evaluate disease severity has certain shortcomings,
Submitted May 20, 2013. Accepted for publication Sep 05, 2013. due to the effects of age, body weight, and defect shape on the
Available at www.jthoracdis.com
diameter (4). Recently, the ratio between VSD area and body
ISSN: 2072-1439 surface area has been considered a more objective reflection
© Pioneer Bioscience Publishing Company. All rights reserved. of disease severity, providing an important reference for both
Journal of Thoracic Disease, Vol 5, No 5 October 2013 601

prognosis and treatment efficacy (5). There is a significantly were recruited from the First Affiliated Bethune Hospital,
different spontaneous closure rate between the patients with ratios Jilin University, between January 2004 and December 2011.
either smaller or larger than 0.5 cm2/m2 (6). However, the method Enrollment criteria included: (I) age ranging from 0.5 to
to calculate the ratio is cumbersome to achieve in clinical practice, 4.5 years, and (II) isolated pmVSD. Exclusion criteria were:
due to difficulties in obtaining the body surface index directly. This (I) complications of VSD affecting the systemic-pulmonary
has resulted in restricted application of the ratio. circulation ratio, such as VSD with membrane aneurysm and
Quantitative assessments of the cardiac shunt have been used aortic valve prolapse; (II) other types of VSD, such as mVSD,
to indicate VSD severity. The systemic-pulmonary circulation sVSD, and mixed-type VSD; (III) VSD as a component of other
ratio (Q P/Q S ratio), left ventricular end diastolic diameter congenital heart defects (such as Tetralogy of Fallot); (IV)
(LVDd) and left ventricular fractional shortening (FS) have all VSD combined with other serious congenital disorders such as
been evaluated, with the most widely used method being the cerebral palsy; (V) VSD combined with other serious diseases of
QP/QS ratio (5). Catheterization is an invasive and expensive other systems, such as leukemia.
technique that is difficult to apply in the clinical setting, even Healthy subjects (n=987) having normal findings on physical,
though it is regarded as the gold standard for measuring electrocardiography, and echocardiography examinations served
VSD severity (7). With the development of color Doppler as the control group. The two groups were age- and gender-
echocardiography, the measurement of QP/QS ratio has gradually matched. The study was approved by the Ethics Committee of
become recognized as a useful method to determine the severity the First Affiliated Bethune Hospital, Jilin University and all
of VSD (5,6). Conventional ultrasonic pulse wave Doppler subjects gave informed consent.
(PWD) is the most common method to measure the Q P/QS
ratio clinically, but this technique is limited due to measurement Subgroups
errors that may be related to its relatively complex operation.
Other methods to measure the cardiac shunt have proven to be The pmVSD group was subdivided into four groups, according
insufficient (8,9). to age: 0.5 to <1.5 y, male n=160 cases and female n=139 cases;
Therefore, there is a clear need to find a simpler method to 1.5 to ≤2.5 y, male n=151 cases and female n=129; 2.5 to <3.5 y,
assess VSD severity, and this measurement should overcome male n=122 and female n=105; and 3.5 to <4.5 y, male n=100,
the effects of age, body mass, and defect shape. Because VSD is female n=81.
a congenital heart disease with a persistent left-to-right cardiac Each age group was further subdivided into four groups
shunt, we hypothesized that the ratio between the diameter of according to DVSD/DAR ratio: (I) 1/5≤ DVSD/DAR <1/4; (II) 1/4≤
the defect and the aortic root diameter (DVSD/DAR) would be a DVSD/DAR <1/3; (III) 1/3≤ DVSD/DAR <1/2; and (IV) 1/2≤
more suitable variable to accurately reflect the degree of shunted DVSD/DAR <2/3.
blood and the severity of VSD.
Based on the location of the defect, VSD is divided into four Height and weight measurements
types: perimembranous VSD (pmVSD), muscular VSD (mVSD),
subarterial VSD (sVSD) and mixed-type VSD. The hemodynamics Measurements were made according to the method of normal
vary among the different types, and thus in different treatments physical development research for children 0-6 years of age. This
and prognoses. The pmVSD is the most common type, so method was provided by the Chinese Ministry of Health.
children with isolated pmVSD were enrolled in the present study.
Accordingly, we calculated the ratio of DVSD/DAR by color Doppler Color Doppler echocardiography
echocardiography in children with pmVSD and determined its
relatedness to the children’s subsequent growth and infection Two-dimensional echocardiograms were obtained with
issues. We evaluated whether this ratio could predict the degree of a TOSHIBA SSH-880CV echocardiography ultrasound
shunted blood and the severity of VSD. (TOSHIBA, Japan). One operator, who was blinded to the
This procedure is easy to perform, since the diameters of the different groups, acquired and analyzed all of the images. A
pmVSD and aortic root are easy to measure. This makes the new 3 MHz frequency probe was placed in the three to four left
method more useful in clinical practice. parasternal intercostals under supine or left lateral position. All
measurements were made according to American Society of
Methods Echocardiography guidelines (10). Three consecutive cardiac
cycles were measured, and average values were obtained.
Patients The following parameters were derived:
(I) The pmVSD size: the minimum diameter of pmVSD at
Patients with isolated pmVSD of either gender (n=987) left ventricular long axis section, the maximum diameter at aorta
602 Liu et al. Predicting VSD complications

short axis apical four -chamber and apical five-chamber section, Body weight decreased in the C and D DVSD/DAR ratio groups
the defect size was calculated as follows, (maximum diameter +
minimum diameter)/2. Compared with the control group, the average body weights​​
(II) The QP/QS ratio: measurement of diameter and velocity of groups A and B were not different, while body weights in
time integral of aortic root and the main pulmonary artery at groups C and D were decreased for all ages (all P<0.05; Table 1).
sections of left ventricular long axis and aortic short axis, QP/ Compared with group C, body weights for group D were lower
QS ratio was calculated as follows: (πr12 V1)/ (πr22 V2); r1, radius in the 1 year subgroup (P<0.05).
of the main pulmonary artery; r2, radius of the aortic root; V1,
velocity time integral of the main pulmonary artery; and V2, Respiratory tract infections increased in the C and D DVSD/
velocity time integral of aortic root. DAR ratio groups
(III) The number of pmVSD, location, color Doppler, and
blood flow crossing the septum. Compared with the control group, infection scores of groups
(IV) DVSD/DAR , ventricular septal defect size/diameter of A and B were significantly increased only in the 1 year age
aortic root. subgroups (both P<0.05; Table 1). Infection scores of each
age subgroup of groups C and D were increased (all P<0.05).
Follow-up investigation Moreover, infection scores of group D were higher than group C
in the 1 year subgroup (P<0.05).
Follow-up questionnaires were used, in addition to the
following: (I) initial diagnosis date; (II) the echocardiography QP/QS increased in the C and D DVSD/DAR ratio groups
report, including pmVSD number, size, location, aortic root
diameter, and complications of a membranous ventricular Compared with the control group, QP/QS of each age subgroup
septal aneurysm; (III) measurements of height and weight; were increased as the DVSD/DAR ratio increased (all P<0.05;
and (IV) details on complications such as the occurrence of Table 1). QP/QS of group C and D were higher than group A and
respiratory tract infections, including details on the duration group B (both P<0.05; Table 1). Moreover, QP/QS of group D
and treatment. was more than 2.

Lower respiratory tract infection score evaluation methods Relationship between DVSD/DAR ratio and measurements of
growth, development, and infection rates
Lower respiratory tract infections were evaluated by determining
pulmonary blood flow. We adopted a semi-quantitative method By linear regression analysis, the DVSD/DAR ratio positively
that recorded respiratory tract infections and developed a scoring correlated with QP/QS and infection scores (Table 2). Further,
system. The scoring rules were: bronchitis, 1 point; pneumonia, the DVSD/DAR ratio negatively correlated with height and body
2 points; Killip II heart failure, 1 point, Killip III or IV heart weight (Table 2).
failure, 2 points; and infections lasting > two weeks, 1 point.
Discussion
Statistics
A congenital heart defect is defined as an abnormality in
Data are expressed as mean ± standard deviation. One-way cardiovascular structure or function that is present at birth,
ANOVA was used for statistical analysis where appropriate. even if it is discovered much later (11). Ventricular septal
A linear regression analysis was used to assess relationships defects represent the most common type and accounts for 25%
between DVSD/DAR and other measurements. All statistical of all congenital heart disease. Although VSDs can undergo
analyses were performed by software SPSS 17.0, and P<0.05 was spontaneous postnatal closure, frequent outcomes include growth
considered statistically significant. retardation, recurrent infections, heart failure, and death (12).
Ventricular septal defect size is a major factor affecting prognosis
Results and is usually assessed by measuring the diameter of the defect (4).
VSD is divided into three types, according to the defect
Height decreased in the C and D DVSD/DAR ratio groups diameter: a small VSD has diameter <5 mm; a medium VSD
has diameter ≥5 and <10 mm; and a large VSD has diameter ≥
Compared with the control group, the average heights ​​of groups 10 mm (13). However using simple VSD diameter to evaluate
A and B were not different, while heights in groups C and D were disease severity has certain shortcomings, due to the effects of
decreased for all ages (all P<0.05; Table 1). age, body weight, defect shape, and defect type. Other methods,
Journal of Thoracic Disease, Vol 5, No 5 October 2013 603

Table 1. Heights, Body weights, Infection scores, QP/QS ratios in the age groups, divided by DVSD/DAR ratios.
Age (y) n Height (cm) Weight (kg) Infection score QP/QS divided by DVSD/DAR
Control 1 299 71±11.9 9.51±0.90 0.55±0.69 1.00±0.27
2 280 8 ±4.1 11.89±0.93 0.40±0.60 1.00±0.23
3 227 96±3.1 13.89±1.17 0.30±0.57 1.00±0.26
4 181 102±3.9 15.26±1.6 0.30±0.57 1.00±0.19
A 1 87 71±3.8 9.50±1.52 2.14±1.42 1.15±0.14
2 83 86±7.1 11.78±1.49 1.25±1.18a 1.20±0.12
3 70 95±7.1 13.30±1.51 0.33±0.62 1.11±0.20
4 60 101 ± 3.7 14.70±1.29 0.45±0.69 1.12±0.14
B 1 74 71±11.9 8.64±1.77 2.22±1.64b 1.40±0.30
2 70 84±5.4 11.55±1.72 1.69±1.49 1.51±0.06
3 62 95±5.4 13.12±1.14 0.92±1.08 1.44±0.10
4 53 100±5.8 14.57±1.21 1.00±1.12 1.53±0.23b
C 1 68 70±11.2b 8.37±2.21b,c 2.60±1.81b,c 1.91±0.55b,c
2 67 83±4.9b 11.03±1.72b,c 2.27±1.42b,c 1.80±0.05b,c
b b b,c
3 55 94 ±7.3 12.84±0.78 1.67±1.73 1.86±0.07b,c
4 40 99±6.5 13.95±1.62b,c 1.50±1.07b 1.89±0.15b,c
D 1 70 68±7.1b 7.23±0.93a,b,c,d 2.94±1.69a,b,c 2.37±0.20a,b,c
b,c a,b,c a,b,c
2 60 82±5.7 10.36±1.21 2.60±1.78 2.17±0.13a,b,c
3 40 93±5.9b 12.36±1.19b 2.56±1.949b,c 2.3 ±0.25a,b,c
4 28 98±4.6b 13.76±1.63b,c 1.89±1.45b,c 2.21±0.12a,b,c
Data are presented as mean ± SD kg. A, 1/5≤ DVSD/DAR <1/4; B, 1/4≤ DVSD/DAR <1/3; C, 1/3≤ DVSD/DAR <1/2; D, 1/2≤ DVSD/DAR <2/3. aP<0.05
compared with the B group; bP<0.05 compared with thecontrol group; cP<0.05 compared with the A group; dP<0.05 compared with the C group.

Table 2. Relationship between DVSD/DAR ratio and measurements is similar to a previous study by our team using simple pmVSD
of growth, development, and infection scores. size, where we showed that in pmVSD patients, the size of the
left-to-right shunt and abnormal hemodynamics in the pulmonary
DVSD/DAR ratio
circulation are important factors in the etiology of impaired
1 year 2 years 3 years 4 years
growth (14). Furthermore, we found that growth condition was
Height –0.248 –0.264 –0.232 –0.21
not different to the healthy children in pmVSD patients with DVSD/
Weight –0.472 –0.454 –0.439 –0.442 DAR <1/3. The height and weight values ​​of each age subgroup were
Infection score 0.436 0.432 0.426 0.423 decreased as the DVSD/DAR ratio increased, with the DVSD/DAR ratio
QP/QS 0.949 0.927 0.930 0.925 negatively correlating with both height and weight.
Data are correlation coefficients (R2). Because left ventricular systolic pressure was significantly
higher than the right ventricle in VSD, there exists a left-right
shunt in ventricular leading to systemic circulation insufficiency
including catheterization, determining the VSD area to body and pulmonary circulation congestion (15). Children with
surface area ratio, or determining the QP/QS ratio, are either VSD not only have limited growth but also have serious
invasive or cumbersome to measure in the clinical practice pulmonary disease, such as recurrent incidences of infection
(5,6). Therefore, it is very important to find a more simple in the lung. Increasing evidence supports the concept that
measurement to determine the severity of VSD. pulmonary infection condition reflects pulmonary blood flow
In the present study, height and weight values ​​of each age and susceptibility to infection (16). Accordingly, we studied the
subgroup were decreased as the DVSD/DAR ratio increased, with relationship between DVSD/DAR ratio and pulmonary infection.
the DVSD/DAR ratio negatively correlating with both height Compared with control group, infection scores of groups A and
and weight. This indicates that the DVSD/DAR ratio reflects the B were significantly increased only in the 1 year age subgroup
growth condition of children with pmVSD. The DVSD/DAR ratio and were significantly increased in all subgroups in groups C
negatively correlated with pmVSD, regardless of age. This result and D. Mechanistically, a large pmVSD is more likely to develop
604 Liu et al. Predicting VSD complications

left ventricular-to-right atrial shunt resulting in pulmonary 4. van den Heuvel F, Timmers T, Hess J. Morphological, haemodynamic, and
circulation congestion (17). clinical variables as predictors for management of isolated ventricular septal
Linear regression analysis showed that the DVSD/DAR ratio defect. Br Heart J 1995;73:49-52.
positively correlated with both QP/QS and the infection score. 5. Sivakumar K, Anil SR, Rao SG, et al. Closure of muscular ventricular septal
These results suggested that the DVSD/DAR ratio could reflect the defects guided by en face reconstruction and pictorial representation. Ann
pulmonary infection condition. Moreover pulmonary infection Thorac Surg 2003;76:158-66.
scores were not increased in pmVSD patients with DVSD/DAR <1/3, 6. Shirali GS, Smith EO, Geva T. Quantitation of echocardiographic
compared with the healthy children. Therefore, pmVSD children predictors of outcome in infants with isolated ventricular septal defect. Am
with DVSD/DAR <1/3 had no limitations in growth and normal Heart J 1995;130:1228-35.
pulmonary health. 7. Kurotobi S, Sano T, Matsushita T, et al. Quantitative, non-invasive
The indications for surgical treatment of pmVSD are: age >2 years assessment of ventricular septal defect shunt flow by measuring proximal
with Q P/Q S >2 : 1 or with infection, growth retardation, or isovelocity surface area on colour Doppler mapping. Heart 1997;78:305-9.
other clinical symptoms (indicating the ineffectiveness of drug 8. Mansencal N, Martin F, Farcot JC, et al. Echocardiographic automated
therapy). In the present study, in cases of DVSD/DAR <1/3, cardiac output measurement of pulmonary output and quantification of
pmVSD had little effect on growth and QP/QS did not change intracardiac shunt. Int J Cardiol 2005;104:25-31.
significantly. In this case, surgical intervention was not necessary 9. Kosecik M, Sagin-Saylam G, Unal N, et al. Noninvasive assessment of left-
and follow-up was given regularly. When 1/2≤ DVSD/DAR <2/3, to-right shunting in ventricular septal defects by the proximal isovelocity
indicators of growth were significantly decreased and QP/QS was surface area method on Doppler colour flow mapping. Can J Cardiol
significantly increased. Moreover, the pmVSD was less likely to 2007;23:1049-53.
spontaneously close. Therefore, according to other indicators, 10. Sahn DJ, DeMaria A, Kisslo J, et al. Recommendations regarding
selective surgical intervention should be done to reduce the quantitation in M-mode echocardiography: results of a survey of
further adverse effects of pmVSD. echocardiographic measurements. Circulation 1978;58:1072-83.
In summary, using the DVSD/DAR ratio as a measurement to 11. Hoffman JI, Kaplan S. The incidence of congenital heart disease. J Am Coll
evaluate pmVSD severity may avoid the drawbacks of measuring Cardiol 2002;39:1890-900.
the simple defect diameter. Though its reliability and prognosis 12. Kleinman CS, Tabibian M, Starc TJ, et al. Spontaneous regression of left
value need to be further confirmed, we recommend that DVSD/DAR ventricular dilation in children with restrictive ventricular septal defects. J
ratio can used as one additional reference index to determine the Pediatr 2007;150:583-6.
size of the perimembranous ventricular septal defect. 13. Axt-Fliedner R, Schwarze A, Smrcek J, et al. Isolated ventricular septal
defects detected by color Doppler imaging: evolution during fetal and first
Study limitations year of postnatal life. Ultrasound Obstet Gynecol 2006;27:266-73.
14. Manso PH, Carmona F, Jácomo AD, et al. Growth after ventricular septal
The potential limitation in this study is the small population defect repair: does defect size matter? A 10-year experience. Acta Paediatr
sample size. A multicenter trial with larger sample size and 2010;99:1356-60.
longer-term evaluation is warranted to determine if the 15. Bhatt M, Roth SJ, Kumar RK, et al. Management of infants with large,
independent clinical contribution of DVSD/DAR ratio is a sensitive, unrepaired ventricular septal defects and respiratory infection requiring
noninvasive marker of VSD. mechanical ventilation. J Thorac Cardiovasc Surg 2004;127:1466-73.
16. Bol Raap G, Meijboom FJ, Kappetein AP, et al. Long-term follow-up
Acknowledgements and quality of life after closure of ventricular septal defect in adults. Eur J
Cardiothorac Surg 2007;32:215-9.
This study was partly supported by the key grant from the 17. Wu MH, Wang JK, Lin MT, et al. Ventricular septal defect with secondary
department of science and technology of Jilin Province (20110457). left ventricular-to-right atrial shunt is associated with a higher risk for
Disclosure: The authors declare no conflict of interest. infective endocarditis and a lower late chance of closure. Pediatrics
2006;117:e262-7.
References

1. Paladini D, Palmieri S, Lamberti A, et al. Characterization and natural

Cite this article as: Liu JX, Wang JH, Yang SR, Liu M,
history of ventricular septal defects in the fetus. Ultrasound Obstet Gynecol
Xu Y, Sun JH, Yan CY. Clinical utility of the ventricular
2000;16:118-22.
septal defect diameter to aorta root diameter ratio to
2. Miyake T, Shinohara T, Fukuda T, et al. Spontaneous closure of
predict early childhood developmental defects or lung
perimembranous ventricular septal defect after school age. Pediatr Int
infections in patients with perimembranous ventricular
2008;50:632-5.
septal defect. J Thorac Dis 2013;5(5):600-604. doi:
3. Chen CW, Li CY, Wang JK. Growth and development of children with
10.3978/j.issn.2072-1439.2013.09.05
congenital heart disease. J Adv Nurs 2004;47:260-9.