Journal of Clinical Densitometry: Assessment of Skeletal Health, vol. 15, no.

2, 205e210, 2012
Copyright 2012 by The International Society for Clinical Densitometry
1094-6950/15:205e210/$36.00
DOI: 10.1016/j.jocd.2011.11.001

Original Article

Bone Quality and Nutritional Status in Children With Congenital

Heart Defects
Chico-Barba Laura Gabriela,1,2 Vivanco-Mu~
noz Nalleli,2 Aviles-Toxqui Dalia Patricia,3
Tamayo Juan,3 Rivas-Ruz Rodolfo,2 Buenda-Hernandez Alfonso,1 and Clark Patricia*,2
1

Department of Pediatric Cardiology, National Cardiology Institute Ignacio Chavez, Mexico City, Mexico; 2Clinical
Epidemiology Unit, Childrens Hospital of Mexico Federico Gomez, Faculty of Medicine UNAM, Mexico City, Mexico;
and 3Mexican Committee for the Prevention of Osteoporosis, COMOP, Hipodromo Condesa, Mexico City, Mexico

Abstract
The aim of this study was to evaluate bone quality and nutritional status in children with congenital heart defects
(CHDs) using quantitative ultrasound. A cross-sectional study was designed. A population-based sample of 75 children with CHD (age: 0e6 yr) from the Department of Pediatric Cardiology at the National Cardiology Institute
Ignacio Chavez was compared with 106 healthy children during 2009. Weight and height were determined in
both groups; bone status was measured at the radius and tibia as speed of sound (SOS). Nutritional status was
defined according to the Waterloo and G
omez index. Chi-square test, Students t-test, and analysis of variance
were used to determine the statistical differences. A linear regression analysis adjusted by age, weight, height,
type of CHD, and birth weight was made. Both groups were similar in sex distribution, prematurity, and birth
weight. Acyanotic cardiopathy with increased pulmonary flow was the most frequent (61.3%). Prevalence of malnutrition was higher in CHD group compared with healthy children ( p ! 0.001), and radius SOS was lower in children with CHD compared with healthy children (3484  180 vs 3575  159 m/s, respectively; 95% confidence
interval: 39.8e143; p 5 0.001). A positive correlation was found between CHD and reduced SOS in the adjusted
linear regression model, r2 5 0.455 ( p ! 0.001). Children with CHD have lower SOS radius values compared
with healthy children, suggesting reduced bone quality regardless of the nutritional status.
Key Words: Bone; children; congenital heart defects; nutritional status; quantitative ultrasound.

Therefore, bone mineral accretion during childhood may be

a critical determinant of osteoporosis risk later in life, making
bone growth a fundamental process (2). Several disorders
may be associated with a reduced bone mineral reserve (ie,
store) in children, particularly in patients with chronic
diseases (3).
Children with congenital heart defects (CHDs) may be exposed to an increased risk of skeletal development with low
bone mineral mass due to hemodynamic and metabolic
changes (4). These factors include hypoxia, increased basal
metabolic rate, fatigue on feeding, low nutrient intake, birth
weight (5), reduced intestinal perfusion, intestinal malabsorption, type of CHD (6), type of surgical correction (7), and
protein-energy malnutrition (7e18). Malnutrition, a common
cause of morbidity, is prevalent in more than 30% of children

Introduction
Osteoporosis is a disease characterized by low bone mass
and structural deterioration of bone tissue, leading to bone
fragility and an increased susceptibility to fractures. It is
well known that the disease occurs primarily as a result of
aging, but there is strong evidence that it can also occur as
a result of impaired development of peak bone mass (1).
Received 05/02/11; Revised 11/04/11; Accepted 11/05/11.
*Address correspondence to: Clark Patricia, PhD, Clinical Epidemiology Unit, Childrens Hospital of Mexico Federico G
omez, Faculty of Medicine UNAM, Dr Marquez No. 162, Colonia Doctores,
Delegacion Cuauhtemoc, CP 06720, Mexico City, Mexico. E-mail:
patriciaclark@prodigy.net.mx

205

206
with CHD (19). Patients with increased pulmonary blood
flow, increased metabolic rate, and/or pulmonary hypertension are more prone to develop malnutrition and show growth
retardation (20). Nevertheless, some studies show that if adequate calories are provided and early corrective surgery is
performed, normal growth potential may be attained in children with most cardiac malformations (21).
Quantitative ultrasound (QUS) has recently been used as
a noninvasive method of estimating bone mass accretion at
the peripheral skeleton anatomic sites (mid-radius, tibia,
and phalanxes). Broadband ultrasound attenuation (BUA)
and speed of sound (SOS) attenuation in bone are measured
using QUS devices. BUA values depend on the trabecular
bone and provide structural information. SOS reflects the
combination of mineral density, maturation phase of the mineral matrix (primary or secondary mineralization), architecture, elasticity, and finally bone strength (22). Some studies
have demonstrated that a reduced value of SOS is associated
with a reduced bone mass accrual status in children with
growth disturbances or disorders affecting bone health; so
QUS can identify a population of children with an increased
risk of bone fragility (3).
Because growth retardation is linked with bone accretion,
this study was designed to evaluate the bone quality and nutritional status in a group of children with CHD compared
with healthy children using bone attenuation of ultrasonic
waves (SOS) to determine differences between the groups
as well as differences between the different types of CHD
and its association with nutritional risk factors.

Materials and Methods

Seventy-five children diagnosed with CHD (age: 0e6 yr) attending a scheduled visit at the Department of Pediatric
Cardiology at the National Cardiology Institute Ignacio
Chavez were compared with healthy age- and sex-matched
controls from a nearby nursery school. Children with conditions
related to growth retardation were excluded (ie, those with
Down, Marfan, Noonan, and Turner syndromes; VACTERL
association, such as vertebral anomalies, anal atresia, tracheoesophageal fistula, esophageal atresia, or renal anomalies).
None of the children with CHD have undergone any surgical
procedure to evaluate if CHD per se affects the bone status.
Children with CHD were categorized into 1 of 4 groups according to the cardiac malformations: (1) acyanotic with increased pulmonary flow (AIPF), (2) acyanotic with normal
pulmonary flow, (3) cyanotic with increased pulmonary flow
(CIPF), and (4) cyanotic with decreased pulmonary flow
(CDPF) (23).
The following demographic and anthropometric parameters were evaluated in both groups: Prematurity (defined as
gestational age of 36 wk and less) and type of cardiac defect
were collected from clinical records; drug prescription information was obtained interviewing the parents. Nutritional
status was defined according to the National Center for
Health Statistics charts for weight/age, height/age, and
weight/height indexes using the Waterloo and G
omez cutoff
Journal of Clinical Densitometry: Assessment of Skeletal Health

Laura Gabriela et al.

points (24,25). In children younger than 24 mo, height was
measured using an infantometer (1-mm precision); children
aged older than 25 mo were assessed using a Seca stadiometer
(1-mm precision). Weight was measured with a Seca baby
scale (0.5-g precision, Seca GmbH & Co., Hamburg,
Germany) and Tanita digital scale (0.1-kg precision, Tanita,
Arlington Heights, Illinois) for children younger than 24 mo
and older than 25 mo, respectively.
A QUS device (Sunlight Omnisense 7000P, Sunlight Medical, Tel Aviv, Israel) was used for bone assessment (in vivo
variation coefficient of 0.3-0.4% reported on the manufacturers manual). Measurements were performed at the distal
third of the radius and midshaft tibia using a single type of
ultrasonic probe. Measurements were done by 2 previously
standardized operators (r 5 0.94).
The participation of the children was obtained according to
institutional standards and approved protocols for research involving human subjects. Written informed consent was obtained for each participant from his or her legal guardian.
The study protocol was approved by the Ethics and Investigation Committee of the National Cardiology Institute.
To study the possible SOS changes in bone tissue accrual
associated with CHD, each patient was matched with
a healthy control subject based on age and gender. SOS differences were then calculated, and the resulting mean value
between the 2 groups was assessed by conducting a t-test.
Chi-square test and Students t-test were done to determine
the statistical differences in every variable. A linear regression analysis adjusted for age, prematurity, and nutritional
status was made to find a relation between lower bone mass
levels and CHD. All statistical analyses were made using
SPSS version 16.0 for Windows (SPSS Inc., Chicago, IL).

Results
Among the 181 eligible healthy patients and 145 children
with CHD, 106 (97.2%) and 75 (51.7%), respectively, agreed
to participate in the study (May to September 2009). QUS was
performed in 173 (99.4%) patients and 179 (96.1%) healthy subjects at radius and tibia. Missing values in radius measurement
were due to positioning difficulties of the age group.
The CHD group comprised 44 females (58.7%), median age
was 37.4  19 mo (median  standard deviation [SD]), and average birth weight was 2.920  0.520 kg (median  SD); 6 of
these children (9.4%) were premature (Table 1). There was
no statistical significance in birth weight and prematurity.
There were 54 females in the healthy group (50.9%); their median age was 43.8  16.6 mo (mean  SD).
The AIPF group was the most prevalent in this sample
(n 5 46, 61.3%), followed by the CDPF group (n 5 15, 20%).
The most common CHDs were ventricular septal defect
(n 5 40, 53.3%), patent ductus arteriosus (n 5 22, 29.3%), aortic stenosis (n 5 13, 29.3%), and pulmonary stenosis (n 5 10,
13.3%).
Prevalence of malnutrition was higher in children with
CHD compared with the healthy group (42.7% vs 2.8%,
p ! 0.001). A normal nutritional status was found in most
Volume 15, 2012

Bone Quality and Nutritional Status in Children With CHD

Table 1
Baseline Characteristics
Children with
CHD, n 5 75
(41.6%)

Healthy children,
n 5 106 (58.4%)

Sex
Male
Female

31 (41.3)
44 (58.7)

52 (49.1)
54 (50.9)

0.304

Pregnancy
Preterm
Term

6 (9.4)
58 (90.6)

13 (12.4)
92 (87.6)

0.374

Nutritional status
Normal
Malnutrition
Obese

41 (54.7)
32 (42.7)
2 (2.7)

86 (81.1)
3 (2.8)
17 (16)

Characteristics

Cardiac defect
None
AIPF
ANPF
CIPF
CDPF

was r2 5 0.455 ( p ! 0.001). No differences in SOS values

were found in weight, height, and birth weight ( p 5 0.814,
0.861, and 0.090, respectively). Age and type of CHD showed
statistical significance for SOS values in radius ( p 5 0.003
and 0.010, respectively) (Table 3). SOS values were increasing according to age in all children.

Discussion

105 (100)

46 (61.3)
5 (6.7)
9 (12)
15 (20.0)
Mean (SD)
Birth weight (kg) 2.920 (0.520)
Age (mo)
37.4 (19)

Mean (SD)
3.099 (0.429)
43.8 (16.6)

0.014
0.018

SOS (m/s)
Radius
Tibia

3575 (159)
3497 (117.7)

0.001
0.131

3484 (180)
3468 (138.8)

207

Abbr: CHD, congenital heart defect; AIPF, acyanotic with increased pulmonary flow; ANPF, acyanotic with normal pulmonary
flow; CIPF, cyanotic with increased pulmonary flow; CDPF, cyanotic
with decreased pulmonary flow; SD, standard deviation; SOS, speed
of sound.

of the children in both groups (CHD: n 5 41, 54.7% and

healthy: n 5 86, 81.1%).
SOS in radius was lower in the CHD group (3484  180
m/s) than in healthy children (3575  159 m/s; 95% confidence
interval: 39.8e143; p 5 0.001). SOS in tibia was also lower in
children with CHD, but the difference was not significant
( p 5 0.12), as shown in Fig. 1. SOS increased with age in
both groups of children.
Regardless of CHD, SOS differences were significant for age
and nutritional status. Malnourished children had lower SOS
than children with a normal nutritional status for both measurement sites ( p ! 0.001). When comparing SOS in healthy children and type of cardiac defects, we found that the most
affected group is CIPF ( p 5 0.003). Differences in SOS values
were not significant among females and males either for radius
or for tibia ( p 5 0.27 and 0.33, respectively) (Table 2).
A linear regression model adjusted for age, weight and
height at baseline assessment, type of CHD, and birth weight
was performed. According to this model, the correlationadjusted association between CHD and reduced SOS at radius
Journal of Clinical Densitometry: Assessment of Skeletal Health

To our knowledge, this is the first study that reports bone

mass determination using SOS attenuation values in bones
of young children with CHD compared with healthy matched
controls. This is a preliminary study.
We assessed bone mass using QUS, a device for bone tissue/structure assessment. QUS is very convenient for this
kind of population because it is portable and radiation free,
making it safe for use in young children; earlier studies
have shown it to be useful in identifying bone abnormalities.
A previous study by Witzel et al (26), with a smaller sample than ours, showed no association between bone development and CHD in adolescents and young adults using
peripheral quantitative computed tomography. In our study,
sample size showed sufficient statistical power (80%) to demonstrate differences between both groups of study in SOS
values.
Results show that radius SOS is lower in CHD group than
in healthy children. There are two ways by which bone mass
can be affected in children with CHD. The first is through nutritional status. Studies have shown that malnutrition affects
bone development; nevertheless, it has been shown that if
adequate calories are provided, normal growth potential
may be fulfilled (21). The second way is represented by hemodynamic and metabolic changes, which include hypoxia
and increased basal metabolic rate. Hypoxia leads to reduced
intestinal perfusion, resulting in intestinal malabsorption of
nutrients, including calcium and vitamin D, which are fundamental in bone development (5,6). In contrast to healthy children, most children with CHD are affected by malnutrition

Fig. 1. SOS in radius and tibia per group. SOS, speed of

sound; CHD, congenital heart defect.
Volume 15, 2012

208

Laura Gabriela et al.

Table 2
Difference in SOS per Variable
SOS radius (m/s)

SOS tibia (m/s)

Variable

Mean (SD)

Mean (SD)

Sex
Male
Female

78
95

3523 (177)
3552 (169)

0.27

83
96

3495 (148)
3476 (105)

0.333

Pregnancy
Preterm
Term

18
143

3466 (212)
3551 (164)

0.048

18
149

3458 (138)
3488 (126)

0.363

Nutritional status
Normal
Malnutrition
Obese

124
31
18

3556 (164)
3423 (184)
3617 (124)

0.000
0.311

126
34
19

3501 (117)
3416 (144)
3503 (124)

0.001
0.998

Group
Healthy children
Children with CHD

105
68

3575 (159)
3484 (180)

0.001

105
74

3497 (117.7)
3468 (138.8)

0.131

3501
3459
3365
3520

0.002
0.115
0.552
0.003
0.814

45
5
9
15

3461
3542
3419
3494

0.194
0.493
0.939
0.389
NS

0.000

Cardiac defect
AIPF
ANPF
CIPF
CDPF

42
5
9
12

(172)
(131)
(216)
(178)

0.002

(135)
(156)
(157)
(131)

Abbr: SOS, speed of sound; SD, standard deviation; CHD, congenital heart defect; AIPF, acyanotic with increased pulmonary flow; ANPF,
acyanotic with normal pulmonary flow; CIPF, cyanotic with increased pulmonary flow; CDPF, cyanotic with decreased pulmonary flow; NS,
not significant.

and hypoxia, particularly the CIPF group; consequently, they

could have disturbances in bone mineralization. This is demonstrated in our study showing lower SOS values in the CHD
group and malnourished children.
However, a linear regression model was done, adjusted for
age, weight, height, type of CHD, and birth weight. We consider that difference in age (3 mo) has no clinical impact;
nonetheless, it was included in the model.
Weight and height at baseline assessment were included in
the regression model as they influence bone density. It is well
documented that prematurity has an effect on bone status; in
Table 3
Linear Regression Model
Unstandardized coefficients
Variable
Age (mo)
Weight at baseline (kg)
Height at baseline (cm)
Birth weight
Type of CHD

b
6.201
1.756
0.648
37.874
25.124

Standard error

2.062
7.441
3.7
22.218
9.662

0.003
0.814
0.861
0.090
0.010

Abbr: CHD, congenital heart defect.

Radius is the dependent variable.
Journal of Clinical Densitometry: Assessment of Skeletal Health

our study, prematurity was directly recorded from the mother,

leading information to recall bias, and so we decided to include birth weight as an indicator of prematurity. Regarding
the presence of CHD, type of CHD was included in the analysis as an indicator of severity of the disease. As the children
with CHD were all considered surgically untreated, age at
baseline assessment could be considered as duration of the
disease; this was one of the most important predictors for
bone status in the regression model analysis.
The model accounts for 45% of the variance of low SOS
values. Specifically, age and type of CHD have more influence on low SOS values ( p 5 0.003 and 0.010, respectively).
The impact of age on SOS scores is explained by bone accretion during childhood; bone mass rises as age increases until
peak bone mass is reached during adolescence. SOS is negatively affected by the presence of CHD as shown in the linear
regression model and when compared with healthy children
using Students t-test. In the analysis model, weight, height,
and birth weight have no influence on SOS values. Sex was
not included in the model because there are no sex differences
on bone accretion during childhood until puberty.
These results encourage clinicians to include bone assessment in children with CHD to achieve an adequate bone development and normal growth. Providing an early and
adequate nutritional therapy (with a correct amount of nutrients important for bone development, such as calcium and
Volume 15, 2012

Bone Quality and Nutritional Status in Children With CHD

vitamin D) as well as a timely palliative/corrective surgery
may help to prevent bone disturbances.
Although reference SOS values are available in healthy Israeli children (27), the normal reference values for SOS attenuation in the normal Mexican population are not available;
therefore, a control group was required to have a good comparative group.
This study has certain limitations. In some cases, birth
weight or gestational age data were not available on the medical records and/or family recalls. Although dual-energy
X-ray absorptiometry measurement is the gold standard for
adults, there is much controversy in its use in children;
some clinical factors described in the Introduction section
(such as hypoxia, basal metabolic rate, nutrient intake, and intestinal perfusion) were not included in the present analysis
because they were not described in detail in the clinical records. Although they may have an influence on bone development, we could not explore their impact. However, the factors
we included in the study are reported to have an impact on
bone accretion (age and prematurity), and their impact was
shown in the analysis. Further studies with a wider analysis
of clinical factors are needed to determine the mechanisms
by which bone accretion may be altered.
Regarding tibia results, the sample size might have been
too small to achieve statistical power. We selected QUS because it is a noninvasive, portable, and radiation-free method
that is safe for use in children.
In conclusion, in this preliminary study, we demonstrate
that children with CHD have lower SOS values compared
with healthy children, suggesting reduced bone quality regardless of nutritional status although malnutrition is present
in most of the children in the CHD group.
In our review of literature regarding bone status and children with CHD, only 1 cross-sectional study was found, and
there is no evidence of longitudinal studies. No previous
studies describing bone status using QUS in this group of
children were found. Further longitudinal studies are needed
to assess bone health after a surgical procedure. Special
medical attention in this vulnerable group of children is required so as to assure optimal bone development and therefore prevent bone accretion disturbances in adolescence and
adulthood.

Acknowledgments
Funding was received from the Science and Technology
National Council (CONACYT).

References
1. World Health Organization. 2003 Prevention and management
of osteoporosis: report of scientific group. WHO Technical
Support Series. World Health Organ Tech Rep Ser 921:
1e164.
2. Tau C. 2006 Densitometra osea en pediatra [Bone densitometry
in pediatrics. Updates on osteology]. Actualizaciones en
Osteologa 2(1):26e28.

Journal of Clinical Densitometry: Assessment of Skeletal Health

209

3. Baroncelli GI. 2008 Quantitative ultrasound methods to assess

bone mineral status in children: technical characteristics, performance, and clinical application. Pediatr Res 63:220e228.
4. Wheat J. 2002 Nutritional management of children with congenital heart disease. Nutr Bytes 8(2):1e5.
5. Vaidyanathan B, Reshma R, Sarala DA, et al. 2009 What determines nutritional recovery in malnourished children after
correction of congenital heart defects? Pediatrics 124(2):
294e299.
6. Da Silva VM, de Oliveira Lopes M, de Araujo TL. 2007 Growth
and nutritional status of children with congenital heart disease.
J Cardiovasc Nurs 22:390e396.
7. Buenda A, Calder
on-Colmenero J, Pati~
no E, et al. 2004 Secuencia de estudio en el ni~
no con cardiopata congenita [Sequence of
assessment in children with congenital heart disease]. In: PAC
Pediatra I. Academia Mexicana de Pediatra. Intersistemas
Editorial, Mexico, 507e605.
8. Velasco C. 2007 Nutrici
on en el ni~
no cardi
opata [Nutrition in
children with congenital heart defects]. Colombia medica
38(Suppl 1):50e55.
9. Olivares JL. 2003 Nutrici
on en el ni~
no con cardiopata congenita [Nutrition in children with congenital heart diseas]. In:
Nutrici
on en pediatra [Nutrition in pediatrics]. Bueno M,
Sarra A and Perez-Gonzalez JM, eds. Madrid, Spain: Ergon,
415e419.
10. Toussaint G, Garca-Aranda JA. 2001 Desnutrici
on energeticoprotenica [Protein-energy malnutrition]. In: Nutriologa Medica
[Medical nutriology]. Casanueva E, Kauffer-Horwitz M,
Perez-Lizaur AB and Arroyo P, eds. Mexico: Editorial Medica
Panamericana, 212e242.
11. G
omez F. 2003 Desnutrici
on [Malnutrition]. Salud P
ublica de
Mexico 45(Suppl 4):s576es582.
12. 2001 Hormona paratiroidea, calcitonina, metabolismo de calcio
y fosfato, vitamina D, huesos y dientes [Parathyroid hormone,
calcitonin, calcium and phosphate metabolism, vitamin D, bones
and teeth]. In: Tratado de Fisiologa Medica [Textbook of medical physiology]. Guyton AC and Hall JE, eds. Mexico:
McGraw-Hill Interamericana, 1081e1100.
13. 2002 Control hormonal del metabolismo del calcio y la
fisiologa del hueso [Hormonal control of calcium metabolism
and bone physiology]. In: Fisiologa medica [Medical physiology]. Ganong W, ed. Mexico: Manual Moderno, 417e432.
14. 2008 Bone mineral acquisition in utero and during infancy and
childhood. In: Osteoporosis. Marcus R, Feldman D and Nelson D,
eds. San Diego, CA: Elsevier Academic Press, 705e723.
15. Thompson-Chagoyan OC, Reyes-Tsubaki N, Rabiela-Barrios OL,
et al. 1998 Estado nutricio del ni~
no con cardiopata congenita
[The nutritional status of the child with congenital cardiopathy]. Archivos del Instituto de Cardiologa de Mexico 68:
119e123.
16. Lyon GR. 2007 Nelson textbook of pediatrics. In: Kliegman R,
Behrman R, Jenson H and Stanton B, eds. New York, NY:
Saunders.
17. Malagon I, Onkenhout W, Klok G, et al. 2005 Gut permeability
in paediatric cardiac surgery. Br J Anaesth 94(2):181e185.
18. Park M. 2008 Pediatric cardiology for practitioners. Philadelphia,
PA: Mosby Elsevier.
19. Villass-Keever MA, Pineda-Cruz RA, Halley-Castillo E,
Alva-Espinosa C. 2001 Frecuencia y factores de riesgo asociados a desnutrici
on de ni~
nos con cardiopata congenita [Frequency and risk factors associated with malnutrition in
children with congenital cardiopathy]. Salud P
ublica de Mexico
43:313e323.
20. Shrivastava S. 2008 Malnutrition in congenital heart disease.
Indian Pediatr 45:535e536.

Volume 15, 2012

210
21. Tokel K, Azak E, Ayabakan C, et al. 2010 Somatic growth after
corrective surgery for congenital heart disease. Turk J Pediatr
52(1):58e67.
22. Knapp K. 2009 Quantitative ultrasound and bone health. Salud
publica Mex 51(Suppl 1):S18eS24.
23. Fausse A, Buenda A, Zabal C. 1993 Cardiologa Pediatrica, diagnostico y tratamiento [Pediatric cardiology. Diagnosis and
treatment]. Mexico: Editorial Medica Panamericana.
24. Waterlow J. 1996. Malnutricion Proteico-Energetica, 555.
Washington, DC: PAHO-WHO.

Journal of Clinical Densitometry: Assessment of Skeletal Health

Laura Gabriela et al.

25. World Health Organization. Child Growth Standards, 2007. Available at: http://www.who.int/childgrowth/standards/en/. Accessed
December 9, 2011.
26. Witzel C, Sreeram N, Coburger S, et al. 2006 Outcome of muscle and bone development in congenital heart disease. Eur J Pediatr 165(3):168e174.
27. Zadik Z, Price D, Diamond G. 2003 Pediatric reference curves
for multi-site quantitative ultrasound and its modulators. Osteoporos Int 14(10):857e862.

Volume 15, 2012